Agent Behavior Modeling

Follow this topic Ask about this topic

Momentum

40 papers in the last four weeks, up 67% on the four weeks before. 0.5% of all new papers.

Jul 6Week of Sep 21

Latest papers 256

Jun 28, 2026cs.AI

A Cognition-Emotion-Personality Framework for Modeling Human-Like Awareness and Behavior in Emergency Evacuations

Agent-based evacuation simulations are widely used to study crowd behavior during emergencies, but many models rely on assumptions such as perfect event awareness, complete exit knowledge, and fully rational decision-making. This paper presents an extended evacuation framework that integrates cognitive, emotional, social, and personality-related mechanisms into a unified model of human behavior under uncertainty. The framework incorporates a dynamic event-awareness mechanism based on a continuous Event Certainty Level, a memory-based representation of exit knowledge subject to acquisition, forgetting, and recall, a continuous fear model in which panic emerges as a high-intensity state, and an OCEAN-based personality representation. Neuroticism is explicitly integrated into the emotional model, influencing fear generation, escalation, social contagion, and recovery. Behavioral heterogeneity is further captured through individualized decision thresholds that affect responses to perceived risk. The framework is evaluated through simulation experiments examining the effects of spatial familiarity, memory robustness, decision sensitivity, emotional dynamics, and personality variation. Results show that cognitive, emotional, and personality-driven processes substantially influence evacuation dynamics, reducing evacuation efficiency and generating realistic crowd phenomena such as delays, confusion, injuries, and socially influenced behaviors. The proposed framework provides a more realistic representation of human behavior in emergency evacuations and supports systematic investigation of the interactions between cognition, emotion, personality, and crowd dynamics.

Zoi Lygizou, Michalis Zervas, Helena G. Theodoropoulou +3

Athena Research Centre, Xanthi, Greece · Hellenic Open University, Patras, Greece

Personality Agent Behavior Modeling Heterogeneity

Jun 28, 2026cs.RO

Behavior Uncloning: Distilling Mode Redirection into Policy Weights without Inference-Time Steering

Behavior-cloned policies often learn multiple behavior modes from demonstration datasets, including modes that are unsafe or otherwise undesired at deployment. For example, a policy trained on diverse handover demonstrations may learn to pass a knife blade-first. Standard remedies such as data curation and inference-time steering either require access to the original demonstrations for full retraining or add substantial inference-time overhead. To address this gap, we propose MoRE(Mode Redirection), which redirects policy rollouts toward desired behavior modes through a short "uncloning" step. Specifically, MoRE distills the redirection signal from a temporary mode classifier into the policy weights to steer behavior. A retain loss balances this edit by preserving desired-mode competence, allowing the standalone policy to suppress unwanted modes with zero inference-time overhead. Across eight simulated and real-world tasks, MoRE improves the average deployment success rate (SR) by 44 percentage points over the original mixed-mode policy. Among all compared adaptation and steering baselines, MoRE achieves the strongest SR and approaches the filtered-data retraining reference, while preserving task competence and inference speed. MoRE also generalizes across robot policy backbones, including Diffusion Policy and the Pi0.5 VLA, diverse task categories, and real-world deployments.

Hao Wang, Jiuzhou Lei, Dayou Li +5

Texas A&M University · University of Wisconsin · Northwestern University +1

Behavior Cloning Robot Policies Inference-Time Steering Agent Behavior Modeling Competence

Jun 26, 2026cs.RO

Unleashing Infinite Motion: Scaling Expressive Quadrupedal Motion via Generative Video Priors

Quadruped robots have achieved remarkable locomotion, yet their behavioral repertoire remains confined to a few gaits--far from the expressive, companion-like presence long envisioned for them. Attempts to import the humanoid recipe of large-scale motion data have inherited one tacit assumption: that robot motion must first pass through an animal body, making data collection dependent on cooperative animals, reconstruction fragile across species, and retargeting ill-posed across incompatible morphologies. We propose Uni-Mo, a fully automated pipeline that removes the animal from the loop by reframing data scarcity as a generation problem: an LLM proposes motion prompts, a video diffusion model synthesizes the corresponding robot behaviors, and the generated videos are lifted into 3D reference trajectories used to train tracking policies deployed on a real Unitree Go2. To make naively-drifting generations reliably extractable, we introduce an Identity Consistency Loss that enforces appearance coherence across frames. We release Quad-Imaginarium at https://github.com/Amap-Robotics/Quad-Imaginarium, the resulting open-source dataset of 7,488 language-annotated quadruped motions (18.5 hours) spanning acrobatic and performative behaviors. We validate 392 randomly sampled motions on a real Unitree Go2 with a 96.7% deployment success rate, complemented by a 97.6% success rate across the full dataset in simulation.

Youzhi Liu, Li Gao, Yifei Qian +3

Amap, Alibaba Group

Training-Set Long-Tail Motions Legged Robots Agent Behavior Modeling Scaling

Jun 25, 2026cs.CV

Fine-tuning a multimodal large language model for clinician-grade autism behavioral scoring from short home videos

Autism spectrum disorder (ASD) affects 1 in 31 US children, yet median age at diagnosis exceeds four years. Artificial intelligence pipelines that provide quantified diagnosis using easy to access observational data (e.g., home videos) could help with earlier diagnosis, and timely delivery of early treatments. We fine-tuned Gemini 2.5 Pro on 400 clinician-rated home videos with low-rank adaptation, training only on 30 behavioral features previously validated to produce reliable predictions when passed to various ML models. On 99 held-out children (49 ASD, 50 neurotypical), inter-rater reliability with clinicians (per-feature weighted Cohen's kappa) improved by 40% (p<0.001), with 27 of 28 evaluable features improving. As an emergent zero-shot capability, direct ASD diagnosis F1 improved by 53% (p<0.001), matching or exceeding clinician outcomes. Classifier-assisted pipelines using fine-tuned LLM-derived behavioral features matched clinician-scored inputs across all tested pathways and achieved 77% accuracy (95% CI: 68-85%) and an AUC of 86% (95% CI: 78-92%). Fine-tuned multimodal LLMs can serve as scalable behavioral feature extractors for use in autism assessment and diagnosis.

Mohammadmahdi Honarmand, Parnian Azizian, Aaron Kline +9

Department of Mechanical Engineering, Stanford University, Stanford, CA 94305, USA · Department of Pediatrics, Stanford University, Stanford, CA 94305, USA · Department of Biomedical Data Science, Stanford University, Stanford, CA 94305, USA +2

Autism Spectrum Disorder Multimodal Clinical Data Multimodal Large Language Models Model Fine-Tuning Agent Behavior Modeling

Jun 25, 2026cs.SE

Speculative Refinement: A Hybrid Autoregressive Diffusion Decoding Strategy and Its Behavior Across Benchmarks

How should we evaluate generation systems that combine autoregressive (AR) and diffusion decoding? We study this question through Speculative Refinement (SpecRef), a training-free hybrid method that warm-starts a masked diffusion language model from an AR draft using entropy-guided selective masking. Evaluating SpecRef across six benchmarks (HumanEval, MBPP, GSM8K, BBH, ARC-Challenge, HellaSwag) with three distinct evaluation protocols (execution-based pass@1, exact-match, log-likelihood scoring), we surface several findings relevant beyond our specific system: (1) code benchmarks conflate structural discovery with logical correctness: providing a syntactic scaffold lifts accuracy from near zero to over 20% without changing the model, indicating that much of the baseline failure is structural; (2) a refinement tension phenomenon where multi-stage correction degrades already-correct tokens, exposing benchmark saturation ceilings invisible to single-model evaluation; (3) log-likelihood and generative evaluation produce different model rankings for the same model pair, suggesting they measure different capabilities; (4) standard Python post-processing silently breaks code evaluation for non-AR generators. These observations apply to any multi-stage or non-autoregressive generation pipeline and point toward more diagnostic evaluation practices.

Aditi Gupta, Neel Mishra, Kushagra Trivedi +1

IIIT Hyderabad, India · Microsoft

Autoregressive Generation Autoregressive Diffusion Diffusion Language Models Agent Behavior Modeling

Jun 25, 2026cs.RO

Inference-Time Robot Behavior Steering through Physically-Aware Reconfiguration of Task-Structure

A central challenge in deploying learned robot policies is inference-time behavior steering: redirecting a policy at test time to satisfy user preferences not anticipated during training, without retraining. Existing methods fail in two modes: end-to-end methods require fine-tuning or expert-level guidance, while neuro-symbolic methods rely on predefined symbols whose edits can result in logically reasonable but physically infeasible plans. To address this challenge, we propose ReStruct, which builds upon a neural automaton policy that decomposes a visuomotor policy into a high-level state-machine skeleton capturing task structure and a low-level continuous controller represented as a residual policy. Specifically, ReStruct adopts the automaton to represent the preference and incorporates it into the skeleton through a synchronous product, thereby reconfiguring the task structure. With the controller kept frozen, the action priors provided by the skeleton are updated accordingly to enable physically-aware control under a modified task structure. Extensive experiments from simulation and real-world show that ReStruct steers a wide range of preferences, from object-centric specifications to temporal-logic constraints, and after steering surpasses existing methods, exceeding VLA models in both task success and preference-following by up to 25%.

Yiyuan Pan, Hanjiang Hu, Shangtao Li +2

Robotics Institute, Carnegie Mellon University, Pittsburgh, PA 15213, USA

Robot Policies Inference-Time Steering Visuomotor Policy Neural Policies Reconfiguration Robot Systems+2

Jun 24, 2026cs.RO

A System for Fast, Resilient, and Adaptable Loco-Manipulation Behaviors on Humanoid Robots

Humanoid robots could take on physically demanding, hazardous, and repetitive work in spaces built for humans. However, a useful robot for these spaces must coordinate locomotion, whole body motion, perception, contact, and operator supervision. This thesis presents a robot-local, runtime-editable behavior authoring and runtime system. Our system strives to be maximally observable, predictable, and directable following Coactive Design principles developed during the DARPA Robotics Challenge. Our operator interface remains continuously synchronized to the robot for runtime authoring, monitoring, and repair. Our behavior architecture uniquely combines object-centric Affordance Templates, organization and logic inspired by Behavior Trees, and runtime-editable perception through a behavior scene and primitive scene actions. Action primitives build on a whole-body controller that supports moving the arms while walking, and use a concurrent action layering algorithm for speed. The behavior library developed during this work covers more than twenty real-robot task variants, including push and pull doors with knob, push-bar, and lever-handle mechanisms, multi-step exploration sequences, obstacle clearing, and reactive table-to-table manipulation tasks. This behavior system has been deployed on many humanoid robots, such as Boston Dynamics' DRC Atlas, NASA's Valkyrie, IHMC and Boardwalk Robotics' Nadia, Unitree's H1-2, and IHMC's Alex. We evaluate our system across capability, speed, reliability, and speed of behavior creation, adaptation, extension, and combination. Our experiments demonstrate that we can adapt, extend, and combine existing behaviors to create novel loco-manipulation behaviors in minutes or hours. Videos: https://www.youtube.com/playlist?list=PLJK5CTyotYqsfgfnXb-09YNFeBose6uEY.

Duncan William Calvert

Department of Intelligent Systems and Robotics, THE UNIVERSITY OF WEST FLORIDA, HAL MARCUS COLLEGE OF SCIENCE AND ENGINEERING, US

Humanoid Loco-Manipulation Humanoid Robotics Robot Systems Agent Behavior Modeling Resilience

Jun 24, 2026cs.RO

CoStream: Composing Simple Behaviors for Generalizable Complex Manipulation

Long-horizon, contact-rich complex manipulation tasks, such as seating a GPU into a PCIe slot, demand both millimeter high precision and out-of-the-box generalization to new tasks. Existing paradigms struggle to satisfy both: classical pipelines use brittle, task-specific interfaces to achieve high-precision control but require costly pipeline redesigns to adapt to new tasks, whereas monolithic end-to-end policies provide better generalization but lack high precision on complex, out-of-distribution tasks unless retrained with new data. Both paradigms share an implicit assumption: once a manipulation capability is acquired, it must be deployed as a rigid pipeline or monolithic whole, rather than being freely decomposed and recomposed. In this paper, we show that complex manipulation capabilities can emerge naturally from the composition of simple, independent behaviors. Rather than deploying a monolithic policy or a rigid pipeline, we propose CoStream, a framework orchestrating foundation models and diverse sensing modalities into multiple composable core behaviors: a semantic behavior extracting spatial constraints via foundation models; a predictive behavior forecasting trajectories by tracking keypoints in imagined videos; and a reactive behavior providing high-frequency tactile and force corrections. On a shared

SE(3)

interface, these outputs compose by right-multiplication into a single pose command at each control step, executed by a compliant controller. We demonstrate CoStream on 8 real-world tasks spanning everyday manipulation and precision assembly, with the strongest gains in contact-rich assembly and object transfer, and show robust recovery from manual perturbations during execution. Website: https://costream-simple.github.io

Haonan Chen, Yuxiang Ma, Stephen Tian +7

Harvard University · Stanford University · Massachusetts Institute of Technology +1

Real-World Manipulation Tasks Contact-Rich Manipulation Rigid Objects Reactive Agent Behavior Modeling

Jun 24, 2026cs.RO

MPC-Injection: Biasing Off-Policy Locomotion RL Toward Controller-Induced Behavior Basins

Reinforcement learning (RL) for locomotion frequently converges to locally optimal but undeployable behaviors, such as vibrating limbs or scooting on the torso, that maximize return without producing a usable gait. We present MPC-Injection, a low-overhead method that steers RL toward a designer-preferred behavior by inserting transitions generated in the same environment by a model predictive controller (MPC). Unlike reward shaping, MPC-Injection does not require redesigning the task reward, and unlike adversarial imitation learning, it adds no discriminator, no kinematic retargeting, and no auxiliary objective. We analyze how the injected transitions bias the learning, allowing the policy to converge to behaviors that pure RL may fail to reach under simple reward functions. On a 2D walker in simulation and with sim-to-real evaluation on a Go2 quadruped, we show that MPC-Injection produces gaits qualitatively comparable to those of reward shaping and adversarial motion priors. We also show that MPC-Injection can complete a barrel roll that pure RL fails to achieve under the same simple reward and can select between trotting and bounding gaits only through changing the injected MPC data.

Roy Xing, Seyoung Ree, Brian Plancher

Dartmouth College · Harvard University

Locomotion Imitation Learning Off-Policy Learning Cross-Embodiment Retargeting Agent Behavior Modeling

Jun 24, 2026cs.LG

Model Forensics: Investigating Whether Concerning Behavior Reflects Misalignment

A central goal of safety research is determining whether a model is misaligned. Prior work has largely focused on detecting concerning behavior. But behavior alone does not establish misalignment: a concerning action can arise from benign causes such as confusion. This motivates model forensics: investigating whether the action was driven by malign intent. In this paper, we propose a baseline protocol for model forensics consisting of two steps, iterated as needed. First, we read the chain of thought (CoT) to generate hypotheses about what drives model behavior. Second, we make edits to the prompt or environment to test these hypotheses. While the CoT is not always faithful, it is a rich source of unsupervised insight that can guide the collection of more rigorous evidence. To evaluate our protocol, we create a suite of six agentic environments where models exhibit concerning behavior, and apply it to each. We establish that Kimi K2 Thinking takes shortcuts due to a genuine disposition towards low-effort actions, by showing this hypothesis successfully predicts its behavior. Through counterfactual experiments, we show DeepSeek R1 deceives out of a desire to be consistent with a previous instance of itself. Our methods nonetheless leave significant room for refinement. For example, when we test whether Kimi K2 Thinking believes it is violating user intent, we find no evidence of such a belief, but without positive controls we cannot confirm our tests would detect it. Overall, we find our simple protocol provides a strong baseline that we hope future work will improve upon. More broadly, our work is a concrete step in developing the growing field of model forensics.

Aditya Singh, Gerson Kroiz, Senthooran Rajamanoharan +1

MATS · Google DeepMind

Agent Behavior Modeling Digital Forensics Adversarial Evaluation Malicious Agents Thoughts

Jun 24, 2026cs.LG

Compositional Behavioral Semantics for State Abstraction in Reinforcement Learning

State abstraction plays a key role in scaling reinforcement learning to complex but structured systems. In studying such systems, a wide range of behavioral structures have been studied in reinforcement learning, including value functions, invariants, bisimulation relations, and behavioral metrics. However, a general principle for determining what structures are provably preserved under state abstraction is still lacking. In this paper, we present a unified framework for defining and analyzing behavioral structures in reinforcement learning. Our framework provides a compositional way to specify behavioral semantics based on local, one-step descriptions of system dynamics. Using this framework, we establish results showing how behavioral structures can be safely transferred between abstract and concrete systems. We further show how to construct quantitative metrics from logical behavioral semantics with soundness guarantees. Together, these results provide a principled foundation for reasoning about behaviors under state abstraction in reinforcement learning and offer reusable definition and proof principles for a broad class of behavioral structures in reinforcement learning.

Yivan Zhang, Ziyan Luo, Manuel Baltieri

The University of Tokyo · RIKEN AIP · Mila - Quebec Artificial Intelligence Institute +3

Agent Behavior Modeling Similarity and Metrics

Jun 22, 2026cs.AI

Measuring Behavior Portability in Large Language Models

Large language models are increasingly deployed as autonomous decision makers, yet the behavioral mapping they exhibit can vary substantially across decision environments that are payoff-equivalent by construction-environments that share identical payoff-relevant structure but differ in surface presentation. This sensitivity renders suite-based evaluation fragile and raises a fundamental question of behavioral portability: how well does a behavioral mapping learned in one decision environment informative on another that preserves the same underlying incentive structure? We introduce a formal framework to measure this property. Our protocol fits an interpretable behavioral model on data pooled from a set of source environments and evaluates its out-of-sample predictive performance in a held-out target environment, benchmarking against an oracle trained directly on target data. Portability is quantified via a loss-agnostic measure that delivers worst-case bounds on the performance of the induced prediction-action mapping in the target environment. In controlled experiments spanning seven canonical economic decision problems, we document substantial and systematic portability losses, suggesting that behavioral characterizations of LLMs obtained in one decision environment cannot be assumed to transfer reliably to structurally equivalent alternatives.

Tianjia Dong, Nadav Kunievsky, James A. Evans

Knowledge Lab University of Chicago

Behavioral Foundation Models Agent Behavior Modeling

Jun 19, 2026cs.CV

A Smart Classroom Behavior Analysis Framework with a New Highly Congested Classroom Dataset

Student behavior detection is important for intelligent classroom analysis but remains challenging in large-class scenarios due to dense instance co-occurrence, asymmetric occlusion, depth-wise scale variation, and fine-grained semantic degradation in distant targets. Existing classroom behavior datasets and general-purpose detectors are insufficient to characterize and address these challenges. This paper constructs the Highly Congested Classroom Behavior (HCCB) dataset, containing 50,229 student behavior instances across seven categories: reading, writing, heads up, sleeping, looking around, bowing head, and using phone. HCCB provides a challenging benchmark that integrates dense distributions, severe occlusion, scale variation, and fine-grained behavioral semantics. To address these issues, we propose ODER-HSFNet, a YOLO-based detection framework tailored to highly crowded classrooms. At its core, ODER-HSFNet introduces three task-specific innovations: the Occlusion-aware Deformable Edge Rectifier (ODER), which strengthens boundary evidence under occlusion; the Hypergraph-State Spatial Fusion (HSSF) module, which integrates local structure enhancement, state-space contextual modeling, and high-order relation aggregation; and the Occlusion-Calibrated Detection Head (OCDetect), which suppresses low-quality Pre-NMS candidates and reduces false positives from occlusion boundaries and neighboring instances. Experiments on two classroom behavior detection datasets show that ODER-HSFNet outperforms mainstream YOLO-series methods, achieving 60.60%/80.12% mAP50:95/mAP50 on HCCB and 57.36%/74.65% on SCB-D3-S. Ablation studies further verify the effectiveness of the proposed design for highly crowded classroom behavior detection.

Wei Xu, Maoxiang Chu, Yuelong Fan +5

Yolo26 Occlusion Agent Behavior Modeling Temporal Graph Neural Networks

Jun 18, 2026cs.MA

Blame is easier than praise: Measuring off-ball defensive performance in football

The defensive performance of football players is commonly measured through a limited number of actions like tackles and interceptions while their continuous impact through positional behaviour has hardly been studied before. We formulate this problem as an attribution over multi-agent spatiotemporal trajectories without player-level ground truth labels, where event-level changes of expected threat are distributed among individuals. We propose a framework that performs this attribution using player involvement scores calculated from defensive pressure areas (DPAs). By computing role-conditioned baselines within automatically detected team structures, we can determine each defender's expected responsibility for threat created through arbitrary passes. The validity and robustness of this approach are evaluated on a uniquely extensive cross-gender and cross-competition data set, including positional and event data from 64 matches of the men's World Cup, 116 matches of the women's German Bundesliga and 336 matches of the men's German 3. Liga. In the absence of a ground truth, we propose an evaluation protocol that combines multiple relatively weak proxies into robust summary scores. We find a validity score that is improved by around 1 standard deviation compared to the best action-based metric and demonstrate that many popular measures show limited validity. The "blame" for conceding high-value actions shows especially strong correlations with external ratings and market values, making it the first published metric in football to reliably measure positioning errors. All code underlying this work is publicly available to support reproducibility and further research.

Jonas Bischofberger, Runqing Ma, Pascal Bauer +2

Centre for Sport Science and University Sports, University of Vienna, Vienna, Austria · Vienna Doctoral School of Pharmaceutical, Nutritional and Sport Sciences (VDS-PhaNuSpo), University of Vienna, Vienna, Austria · VfB Stuttgart 1893 AG, Stuttgart, Germany +2

LLM Defense Mechanisms Ground Truth Agent Behavior Modeling

Jun 16, 2026cs.LG

Do as the Romans Do: Learning Universal Behaviors from Heterogeneous Agents

Humans often acquire new skills by observing others, since observed behaviors implicitly reveal how to act in an environment. However, observations drawn from a heterogeneous population introduce conflicting behavioral signals, making it difficult to determine which behaviors are worth imitating. We address this challenge with General Reward Inference and Disentanglement (GRID), a social learning method that extracts universally useful behaviors from a heterogeneous population of demonstrators pursuing different goals. GRID decomposes per-agent reward functions into a general reward, capturing behaviors shared across all agents, and specific rewards, capturing individual preferences and objectives. Training exclusively on the general reward provides a new paradigm of generalist pretraining. It yields a generalist agent that internalizes universal environmental competencies, such as safety and basic task proficiency, without the mode-averaging bias that afflicts standard learning from demonstration techniques. This generalist serves as a superior prior for fine-tuning to downstream tasks, including preferences unseen during training. Experiments across a synthetic basis function decomposition, multi-agent Craftax, and a continuous autonomous driving simulator (Highway-Env) confirm that GRID successfully disentangles reward structure in a semantically meaningful way, outperforms standard learning from demonstration baselines, and enables more efficient and stable specialization.

Caleb Chang, Davin Win Kyi, Natasha Jaques +1

University of Washington · NVIDIA

Multi-Agent Reinforcement Learning Agent Behavior Modeling Autonomous Driving Simulation

Jun 16, 2026cs.HC

Towards Understanding and Measuring COGNITIVE ATROPHY in LLM Behaviour

Recent incidents involving LLMs used for mental-health support reveal a critical evaluation gap: surface-level safety scores do not capture how models behave across realistic, emotionally sensitive interactions over time. Existing benchmarks measure knowledge, safety, or static response quality, but miss whether LLM interactions help users keep reflecting, coping, and making decisions themselves. We formalize this missing dimension as COGNITIVE ATROPHY, a process-level behavioural measure in AI-mediated mental-health support distinct from safety and helpfulness. To measure it, we introduce COGNITIVE ATROPHY BENCH, a clinically grounded benchmark built from 1,576 fully human-generated counseling conversations, 15,680 turns, and 42,230 responses from five LLMs. Three clinical and neuropsychology experts developed a 20-attribute schema spanning user context, response behaviour, and global risk flags; six trained clinical reviewers applied it with span-grounded evidence, producing 5,324 reviewer judgments. We further introduce the User-Input Risk Index (UIRI), the Cognitive Atrophy Risk Index (ARI), and trajectory summaries. Across five LLMs, models show a consistent moderate-to-high level of atrophy-aligned behaviour across single and multi-turn settings. While models generally respond to overt safety cues, they adapt less reliably when users seek solutions or decisions. The dominant recurring patterns are directive advice, problem-solving, recommendation responses, topic shifts, and forms of validation that may reinforce dependence rather than reflection. Our work makes COGNITIVE ATROPHY measurable and provides a foundation for auditing model behaviour in sensitive LLM conversations.

Abeer Badawi, Moyosoreoluwa Olatosi, Negin Baghbanzadeh +5

York University, Canada · Vector Institute, Canada · Dalhousie University, Canada +3

Agent Behavior Modeling

Jun 16, 2026cs.LG

SAE Interventions are Unreliable: Post-Intervention Recovery of Suppressed Behavior

Sparse Autoencoders (SAEs) decompose residual-stream activations into interpretable features. Recent latent-space defenses increasingly rely on these decompositions, assuming that identified "unsafe" SAE features serve as actionable handles for monitoring and intervention. In this paradigm, clamping a specific harmful feature is expected to reliably prevent model misbehavior. However, we show that this success may hide a recoverable failure mode: the clamp may block one visible route to a behavior without eliminating the behavior itself. We formulate this vulnerability as post-intervention recovery, a constrained residual-space optimization problem. Starting from the post-intervention residual state, we optimize residual perturbations to recover the pre-intervention behavior while preserving the post-intervention values of the targeted SAE features. Even under a strong threat model where the intervention remains active throughout optimization and generation, recovery remains possible. To rule out that recovery simply undoes the intervention, we use encoder-orthogonal updates for single-layer interventions and the corresponding feature-map Jacobian in the cross-layer setting. Across TPP, unlearning, IOI, and refusal steering experiments, this stress test reveals recoverable behavior despite successful feature-level intervention. Especially in the safety-critical refusal-steering setting, we achieve a 95.8% recovery rate on valid samples while keeping defended-feature relative drift to 0.131, substantially below suffix-based baselines. A recovery-path attribution analysis further localizes this recovery to the SAE reconstruction residual, the component left unexplained by the SAE. These results expose a gap between feature-level control and behavioral completeness: SAE features can support causal intervention, but controlling them does not guarantee control over the underlying behavior.

Mingyue Cui, Linghui Shen, Xingyi Yang

The Hong Kong Polytechnic University

Sparse Autoencoder Features Latent Feature Interventions Suppression Latent Failure Patterns Agent Behavior Modeling

Jun 16, 2026cs.RO

Embodiment Shapes Rolling Behavior in a Multimodal Infant Model

Rolling over is one of the earliest milestones in infant motor development, reflecting the emergence of coordinated, whole-body sensorimotor control. Here, we conduct a computational study of infant rolling using MIMo, a virtual infant embodiment equipped with proprioception and vestibular sensation. MIMo learns supine-to-prone rolls with reinforcement learning. Interestingly, the learned behaviors capture developmental trends and coordination patterns consistent with those reported in real infants, including improved performance and faster execution with age. Our results explain how infant capabilities and constraints can give rise to realistic behaviors in artificial agents, with a particular emphasis on how motor development is shaped by the changing body morphology. This work highlights the role of embodied computational models as a powerful tool for studying sensorimotor development.

Leon Philipp, Francisco M. López, Jochen Triesch

Frankfurt Institute for Advanced Studies, Germany · University of New South Wales, Australia · Goethe University Frankfurt, Germany

Sensorimotor Experience Motion Imitation Embodied Developmental Trajectories Agent Behavior Modeling Multiple-Input Multiple-Output

Jun 16, 2026cs.AI

Dissecting model behavior through agent trajectories

AI agent performance is not just a modeling problem, it is fundamentally a systems problem. The advanced capabilities of models are realized through agent harnesses. Therefore, a gap between model assumptions and harness behavior can easily prevent the model's full capabilities from translating into agent performance. We formalize this as the

intent-execution' gap: the mismatch between what the model intends and what the harness executes, and vice versa. We argue that minimizing this intent-execution gap is as important as other aspects of harness design such as tools and execution loops. To illustrate the impact of this harness-model alignment, we develop a simple and customizable harness called

Simple Strands Agent' (SSA). SSA aims to find the bulk of common patterns which generalize across different model families (such as Claude, Gemini, GPT, Grok, Qwen), as well as a small number of model-specific preferences. We make two contributions: (i) we reproduce or improve on the pass@1 performance reported by diverse model-provider families on popular agentic benchmarks (SWE-Pro, SWE-Verified and Terminal-Bench-2), and (ii) building on an analysis of 138k trajectories generated by SSA, we look beyond the pass@1 numbers which tend to be relatively even across frontier models. By representing agent trajectories in code state-spaces, we observe model-level differences in problem-solving behavior. Finer-grained metrics such as edit frequency, testing activity, and phase-transitions reveal how individual models allocate effort across different stages of autonomous problem solving.

Gaurav Gupta, Vatshank Chaturvedi, Jun Huan +1

AWS AI Labs

Agent Harness Agentic Benchmarks Agent Behavior Modeling Problem Solving Capability

Jun 15, 2026cs.SE

Bistable by Construction: Wall-Clock-Calibrated State Monitors Have No Moment-Detection Regime at Agent Cadence

Runtime monitors for autonomous agents commonly threshold an accumulated internal state - a behavioural baseline, a drift statistic, or, in our prior work, a modelled affective state. We previously reported a State Saturation Trap: threshold-on-state triggers over a continuous affect engine become near-constant alarms on SWE-bench debugging agents (Modgil 2026). A post-release audit found the engine received dt=0 between actions, so its exponential decay never operated: the published trap is a pure-accumulator result. We correct the record (erratum, v2) and treat the flaw as an experiment. The key variable it exposes is whether a monitor's dynamics are calibrated in sample time (per observation, as in CUSUM) or wall-clock time (half-lives in seconds, as in affect models and EMA baselines). On fixed-rate streams these coincide; on agent streams, where inter-action time varies by orders of magnitude, they do not. A pre-registered sweep over uniform intervals (dt in {0..600}s) on 20 trajectories shows the wall-clock level trigger has two regimes: at dt<=1s a constant alarm (20/20; median 18 firings); at dt>=60s silent. Every critical dt lies in (1,30]s. Real agent runs measure latency at median 1.53s (p90 2.33s); real coding cadence sits inside the trap regime, vindicating the empirical finding under a corrected mechanism. The structure is a property of the calibration class, not the engine: a minimal wall-clock accumulator over the raw error stream reproduces the same cliff, while a sample-time CUSUM over the identical stream is exactly dt-invariant (20/20). A rising-edge trigger with hysteresis fires 0-3 times per trajectory in every condition. We conclude that wall-clock-calibrated leaky-integrator monitors admit no regime in which they act as moment detectors on agent streams; transition detection escapes the trap at every cadence, but does not recover human intervention timing.

Manvendra Modgil

Modint Intelligence

Runtime Enforcement State-Tracking Autonomous Agents Deterministic Replay Decision Thresholds Agent Behavior Modeling

Jun 15, 2026cs.SE

Agent trajectories as programs: fingerprinting and programming coding-agent behavior

Benchmark scores tell you what an agent got right; they do not tell you how it got there. In this work, we introduce methods for comparing agents procedurally in different contexts, where the model, tasks, and approaches vary. We compare ten agents and find that they are identifiable by their behavioral habits, which we define as fingerprints: a probe over these procedural signatures attributes an unseen trajectory to the correct agent at 85.7% accuracy, controlling for leakage across tasks. We develop procedural representations for agent problem-solving procedures with an emergent vocabulary induction technique that is meant to be maximally compressive to avoid surface-level variation while being expressive enough to unveil the quirks of the models' patterns. We apply our framework to the software engineering evaluation dataset SWE-Bench to study the structural distinctness of agent trajectories and find that behavior is most similar between models from similar release periods and those that are distilled from one another (e.g., a distilled student model and its teacher have a Jensen-Shannon divergence of 0.25, about half the distance between other model pairs). As more models saturate evaluations, we believe that it will be important to probe model behavior along more holistic dimensions than success rates alone. We introduce ProcGrep, a library for auditing and evaluating agents for how they approach tasks at a procedural level given their traces in a top-down fashion. We believe this work has a range of applications to help developers work with and program coding agents, such as task-aware model routing, agent monitoring, and finer-grained cost analysis.

Hamidah Oderinwale

Work done with support from Taste Labs

Coding Agents Real-Tool Benchmarks Procedural Agent Behavior Modeling Fingerprint

Jun 15, 2026cs.CL

Understanding the Behaviors of Environment-aware Information Retrieval

Recent retrieval-augmented generation (RAG) approaches have demonstrated strong capability in handling complex queries, yet current research overlooks a critical challenge: different retrievers require fundamentally different query formulation strategies for optimal performance. In this work, we present the first systematic analysis of how LLMs can learn to adapt their query formulation strategies for different retrievers via reinforcement learning (RL). Our empirical study reveals that RL effectively teaches an LLM to tailor its queries to specific retriever characteristics. We discover that different retrievers exhibit surprisingly distinct optimal query styles (e.g., descriptive vs. question-like), suggesting strategies learned for one retriever ineffective for another. We further show that performance can be enhanced by incorporating retriever-specific human guidance and by scaling model size. To facilitate learning over multi-retrieval-step trajectories, we introduce a branching-based rollout technique that improves training stability. Our work provides the first empirical evidence and actionable insights for building truly retriever-aware RAG systems. Code and resources are available at https://github.com/LCO-Embedding/Envs-aware-Information-Retrieval.

Ruifeng Yuan, Chaohao Yuan, David Dai +4

1Fudan University · 2Alibaba DAMO Academy · 3Chinese University of Hong Kong +2

Retrieval-Augmented Generation Pipelines Agent Behavior Modeling

Jun 15, 2026cs.RO

Pride and Prejudice: Toward an Information-Theoretic Framework for Mutually Communicative Driver Behavior Modeling

Mixed autonomy driving becomes unsafe and inefficient when autonomous vehicles (AVs) and human-driven vehicles (HVs) misread each other's intentions. We study this problem as implicit mutual communication in lane changes. The proposed framework models how the ego vehicle both expresses its intent and probes the other driver's preference under epistemic uncertainty. It combines a level-k Bayesian persuasion game with virtual features for proactive signaling, information-theoretic rewards for mutual communication, and adaptive weights of communication affordances. We further introduce the Pride-Inquiry (P-I) and Pride-Prejudice (P-P) planes to analyze communication intensity and tendency. The model is calibrated with a Communication-Based Multi-Agent Inverse Reinforcement Learning algorithm (C-MIRL) on the naturalistic NGSIM dataset. Compared with the non-communicative baseline, the proposed model reduces the prediction error of mandatory lane changes by up to 20% while maintaining strong generalization. Driver-In-the-Loop questionnaire scores are positively correlated with the calibrated communication variables, supporting the subjective validity of the model. The learned rewards further show that inquiry and listening affordances contribute more than pride and expression alone, and that inquiry preference varies more strongly across drivers. These results support explicit modeling of mutual communication and epistemic uncertainty in interactive driving.

Tingjun Li, Nan Xu, Shuo Feng +3

State Key Laboratory of Automotive Chassis Integration and Bionics, Jilin University, 5988 Renmin Street, Changchun, 130025, Jilin, China · Department of Automation, Beijing National Research Center for Information Science and Technology, Tsinghua University, Beijing, China · Department of Engineering, Brock University, 1812 Sir Isaac Brock Way, St. Catharines, L2S3A1, Ontario, Canada +1

Persuasion Multi-Agent Reinforcement Learning Semantic-Affordance Inference Agent Behavior Modeling Information-Theoretic Limits

Jun 14, 2026cs.CL

The Truth Stays in the Family: Enhancing Contextual Grounding via Inherited Truthful Heads in Model Lineages

Recent advances in large language models (LLMs) have produced many specialized multimodal LLMs (MLLMs) that share common foundational LLMs, forming distinct model lineages. It remains unclear whether a fundamental behavioral link exists between the foundational LLMs and downstream variants. We investigate this question by quantifying head-level context-truthfulness scores. Across diverse LLM and MLLM lineages, including Vicuna-, Qwen2.5-, LLaMA2-, and Mistral-based models, we find that Truth Scores are strongly preserved within model families, even after instruction tuning or multimodal adaptation. We further show that this inheritance is consistent with attention-head weight preservation, and that context-truthful heads attend to query-relevant evidence. Building on this finding, we propose TruthProbe, a soft-gating strategy that amplifies context-truthful heads while preserving other head contributions. TruthProbe improves contextual truthfulness on HaluEval and reduces multimodal hallucination on POPE and CHAIR, with base-LLM Truth Scores transferring effectively to their fine-tuned LLM and MLLM descendants. Code is available at https://github.com/miso-choi/TruthProbe.

Miso Choi, Seonga Choi, Mincheol Kwon +3

Korea University, Seoul, Republic of Korea · Kakao Mobility Corp., Seongnam, Republic of Korea

Large Language Model Reliability Multimodal Large Language Models Truth Large Language Model Fine-Tuning Contextual Grounding Lineages+1

Jun 14, 2026cs.RO

Can Causal Models Enhance Robot Navigation? Online Causal Adaptation for Real-Robot Navigation

Causality in robotics aims to produce more interpretable and flexible robot behaviours by enabling robots to predict the consequences of their actions; however, deploying causal models with existing systems (e.g., navigation) operating in real environments remains understudied. This paper addresses the challenging problem of transferring causal models in real-robot experiments for a navigation scenario. We study this problem in two ways: (i) using the causal model as an offline evaluation module that predicts the competence of recorded real-robot navigation trajectories and relates it to quantitative navigation performance, and (ii) using the causal model as an online adaptation module that intervenes when the predicted competence of the default navigation is low. We validate our approach in a physical service robot that patrols around corridors. We show that the predicted competence correlates positively with path efficiency, and negatively with path irregularities (suboptimal behaviour). The model predictions also show strong agreement with human annotations (Cohen's kappa value of 0.88). In online experiments, the proposed method improves navigation performance in complex scenarios such as cornering and obstacle avoidance, yielding higher predicted competence and better navigation metrics than the default navigation baseline. In simpler scenarios, where the baseline already performs near-optimally, the causal adaptation provides limited benefit. These results indicate that causal models are particularly effective in enhancing navigation under increased task complexity. Overall, our results demonstrate that causal models developed for behavioural interpretation can be successfully integrated into real-robot navigation systems.

Zhitao Liang, Alex Mitrevski, Emmanuel Dean +1

Division for Systems and Control, Department of Electrical Engineering, Chalmers University of Technology, Gothenburg, Sweden

Robot Navigation Obstacle Avoidance Agent Behavior Modeling

Jun 12, 2026cs.MA

Hierarchical Generative Agents for Simulating Sequential Human Behavior

Complex cognitive, emotional, and social processes shape human evacuations during natural disasters. Accurate modeling and understanding of human behavior in disasters or emergencies can greatly impact the evacuation process by informing more effective planning and resource allocation. However, collecting human data in these situations is very difficult, and existing computational evacuation models assume rational, homogeneous behavior, leading to unrealistic, overly optimistic predictions. To address this gap, we present a simulation framework of sequential human decision-making during an evacuation scenario, introducing cognitively grounded, persona-driven agents. Our framework models evacuation behavior in a grid-based urban environment that evolves over time, capturing fire and other hazards. Human agents are modeled as personas that make sequential decisions in response to environmental stimuli with cognition structured in three levels: high-level evacuation goals, mid-level route reasoning, and low-level navigation. Decision-making is driven by large language models (LLMs) coupled with a cognitive module and calibrated with empirical human evacuation data. We propose a dynamic, stimulus-driven disaster simulation framework that models human evacuation decision-making using persona-conditioned LLM agents and a cognitive hierarchy.

Maria G. Mendoza, Lucas Waldburger, Jin Lee +1

University of California, Berkeley

Cognitive Science Agent Behavior Modeling

Jun 12, 2026cs.RO

TacStyle: Personalizing Tactile Robot Policies using Structured Behavior Representations

Robotic systems that assist humans should be capable of adapting their behaviors to individual user preferences. For instance, users may want a robot arm to adjust the amount of force it applies while folding their laundry or cleaning furniture. Natural language provides an intuitive way for humans to communicate such preferences. Recent progress in language-conditioned robot policies has shown that robots can successfully use language prompts to determine what task to perform. However, extending the same approach to realize how the task should be performed requires detailed labels describing the preferences or styles of trajectories in the task data. Not only is collecting such annotations challenging, but conditioning directly on these labels may also fail to provide fine-grained control over a continuous range of behaviors. For example, it can be difficult to convey the exact force that a robot must apply through abstract instructions like "apply a bit more pressure than before". Therefore, in this work, we propose using language to reason over preferred behaviors instead of directly generating them. We first learn a structured latent representation that organizes user preferences according to differences in the corresponding trajectories. Then, given a preference prompt, we use a foundation model to interpret this latent space and choose a value that produces the desired behavior. Through both simulation and real-world experiments, we show that selecting robot behaviors from an intuitively structured latent space enables more precise adaptation to user preferences while requiring significantly fewer preference labels than language-conditioned policies.

Kevin Robledo, Matías I. Torres Galaz, Kumar Dixhant Rai +3

Department of Computer Science California State University, Northridge · Department of Mechanical Engineering California State University, Northridge

Robot Policies Behavior Tree Robot Systems Tac Agent Behavior Modeling Preference Alignment Learning+2

Jun 12, 2026cs.CL

OdysSim: Building Foundation Models for Human Behavior Simulation

Large language models are increasingly deployed as human simulators for interactive evaluation and social simulation. Yet helpfulness-driven post-training pulls them toward a homogeneous, overly agreeable assistant register, creating a behavioral Sim2Real gap. We present OdysSim, the largest open systematic investigation of behavioral foundation models, i.e., models trained to simulate human behavior at scale. We propose SOUL, a taxonomy of five capability axes (CONV, SS, COG, ROLE, EVAL) that unifies 62 datasets and 23 benchmark tasks under one framework. Specifically, we curate the OdysSim corpus (21.4M interactions, 10B tokens, retrofitted with back-generated social contexts), construct the SOUL-Index benchmark, and develop an end-to-end training recipe combining midtraining, task-specific RL, and expert distillation. The resulting open 8B OSim model ranks first or tied-first on 8 of 23 tasks, outperforming any individual frontier model by this count, with the strongest gains on conversational and social tasks. Its outputs are also more human-like in length, formatting, and word choice, and it transfers zero-shot to out-of-distribution user simulation on

τ

-bench, nearly matching real users on reaction alignment (93.2 vs. 93.5). We further show that LLM-as-judge RL induces reward-hacking patterns, and that our detectors can mitigate them during post-training. Together, our findings suggest that behavioral foundation models require rethinking the LLM training paradigm. We release all artifacts to support future research.

Xuhui Zhou, Weiwei Sun, Weihua Du +6

Carnegie Mellon University, Language Technologies Institute

User Simulation Behavioral Foundation Models Agent Behavior Modeling Post-Training Sim-To-Real Gap Text Corpora

Jun 12, 2026cs.CL

Harsher on Male? Evaluating LLMs on Gender-Asymmetric Moral Framing Across Diverse Conflict Scenarios

Existing studies on gender bias in LLMs have largely focused on stereotypes, occupational associations, or explicit harmful outputs. In this work, we ask whether LLMs apply consistent response standards to the same negative behavior under matched male-actor and female-actor conditions. We introduce GAMA-Bench, a gender-mirrored benchmark of 1,298 scenarios covering intimate relationship and public social conflicts. It constructs gender-neutral misconduct templates through controlled grids and cross-model review, then compiles them into paired first-person prompts with matched actor-gender and role-reference variations. We further design a structured response-framing protocol to measure how models allocate punishment, empathy, escalation, instruction, and blame. Experiments on 10 representative LLMs reveal a consistent male-disadvantaging asymmetry: male actors receive more punitive, escalatory, and blame-centered framing, whereas female actors receive more therapeutic and empathy-oriented framing for the same misconduct. Further analyses show that this pattern persists across model families, scenario tracks, model scale, and explicit thinking-style reasoning. The official code is available at https://github.com/xufeiqiong/GAMA-Bench.

Guangzong Si, Dong Wang, Zhenhao Li +3

University of Science and Technology of China · Eastern Institute of Technology, Ningbo

Moral Reasoning Large Language Model Decisions Large Language Model Alignment Agent Behavior Modeling Hard

Jun 12, 2026cs.CV

HARBOR: Heading Analysis and Reconstruction from Behavioral Observation and Radar

Maritime situational awareness often relies on Automatic Identification System (AIS) transmissions to track vessel movements. However, in operational or conflict scenarios, these data may be unavailable due to signal loss, deliberate deactivation, or intentional spoofing. In such conditions, synthetic aperture radar (SAR) imagery becomes a critical sensing alternative for wide-area maritime monitoring, despite providing only static scene snapshots. This work introduces HARBOR (Heading Analysis and Reconstruction from Behavioral Observation and Radar), a complete pipeline for transforming a single SAR image into predictive motion information without requiring any auxiliary data source at inference time. The method begins with SAR image preprocessing to enhance and segment vessel candidates, followed by automatic detection, size-based classification, and heading estimation using skeleton geometry and local intensity patterns. AIS data are used exclusively during an offline calibration phase to derive vessel-type-dependent motion parameters, which are then applied to generate probabilistic heatmaps of candidate future vessel positions. A case study using real COSMO-SkyMed SAR imagery demonstrates the pipeline on a maritime scene in southern Brazil, showing its ability to extract motion tendencies and generate probabilistic projections of vessel positions in data-denied environments.

Joao P. A. Dantas, Paulo F. Silva Filho, Jelton A. Cunha +1

Institute for Advanced Studies, São José dos Campos, São Paulo, Brazil

Ship Detection Synthetic Aperture Radar Underwater Radar Automatic Identification System Agent Behavior Modeling

Jun 11, 2026cs.AI

Automated reproducibility assessments in the social and behavioral sciences using large language models

Reproducibility in the social and behavioral sciences is typically evaluated by independent researchers who reanalyze the original data to assess whether the published findings can be recovered. However, such approaches are resource-intensive and difficult to scale. Here, we show that large language models (LLMs) can automate reproducibility assessments. Using N = 180 published studies with predefined claims from the behavioral and social sciences, we compare LLM-generated analyses with the original findings. For 11 studies, the LLM pipeline could not produce a viable effect size estimate. For the remaining studies, the LLM reached the same qualitative conclusion as the original study in 80% of cases, and recovered the original effect sizes (using a +/-0.05 tolerance in Cohen's d) in 24% of studies. In a subset with human reanalyses, the LLM reached the same qualitative conclusion as the original study in 95% of studies, similar to human reanalysts (83%), and the LLM recovered the original effect sizes using a +/-0.05 tolerance in 40% of studies, again broadly similar to human reanalysts (28%). Given the current capabilities and limitations of LLMs, the findings show that LLMs can support systematic audits of empirical results rather than substitute expert judgment. As such, LLMs can serve as a scalable screening tool to improve the rigor and reproducibility in empirical research.

Tobias Holtdirk, Pietro Marcolongo, Anna Steinberg Schulten +7

1LMU Munich, Munich, Germany · 2Munich Center for Machine Learning, Munich, Germany · University of Maryland, College Park, USA +1

Reproducibility Effect Sizes Research Automation Agent Behavior Modeling

Jun 11, 2026cs.AI

EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery

LLM-based agents have shown increasing potential in automating scientific discovery. Given an optimizable metric and an execution environment, they can propose, validate, and iterate scientific solutions, and have produced results that outperform human-designed approaches. As model capabilities continue to improve, we argue that the bottleneck for autonomous scientific discovery is shifting from prescribing agent workflows to designing agent environments: the resources, constraints, and interfaces that shape agent behavior. We frame this as environment engineering: building environments that amplify productive behaviors, such as open-ended exploration, systematic artifact management, and inter-agent collaboration, while suppressing harmful behaviors, such as reward hacking and high-friction human oversight. We present EurekAgent, an environment-engineered agent system for metric-driven autonomous scientific discovery. EurekAgent engineers the environment along four dimensions: permissions engineering for bounded agent execution and isolated evaluation; artifact engineering for filesystem and Git-based collaboration; budget engineering for budget-aware exploration; and human-in-the-loop engineering for easy human supervision and intervention. EurekAgent sets new state-of-the-art results on multiple mathematics, kernel engineering, and machine learning tasks, including new state-of-the-art 26-circle packing results discovered with less than $11 in total API cost. We open-source our code and results, and call for environment engineering as a core research direction for developing reliable autonomous research agents.

Amy Xin, Jiening Siow, Junjie Wang +5

Department of Computer Science and Technology, Tsinghua University · School of Information, Renmin University of China

Agentic Discovery Autonomous Agents Agent Behavior Modeling

Jun 11, 2026cs.AI

Nous: An Attempt to Extract and Inject the Cognition Behind Prediction-Market Behavior

As LLM agents proliferate in prediction markets and collective decision-making, they risk a cognitive monoculture: agents built on shared foundation models produce correlated forecasts, and recent measurement finds frontier-model errors correlated at r ~ 0.77. We ask whether human cognitive diversity can be recovered from behavior and transferred to LLM agents. Nous extracts a structured eight-dimension behavioral profile from real Polymarket trading activity and injects it into agents through prompts. Our central finding is a dissociation between the two halves of that pipeline. Extraction works, partially: across 100 wallets, 8 of 14 parameters are temporally stable (split-half ICC >= 0.5, bootstrap CI lower bound > 0.3; contrarian score reaches ICC ~ 0.9); wallets are identifiable from their profiles well above chance (top-1 retrieval 17-22% vs. 1% chance); and two of four pre-specified dimensions rank-correlate with future realized profit out-of-sample, though the correlations do not survive behavioral-confound controls. Prompt-level injection does not measurably transmit it: on a semantic embedding metric, structured injection shows no significant advantage over a length-matched control on any model, and the diversity it induces neither reduces ensemble error correlation nor improves Brier score -- a null that persists across exploratory checks on sampling temperature, profile diversity, and question difficulty. Measuring the prompts themselves locates the compression before the model: the structure-to-narrative translator emits near-uniform prompts whose spread does not track profile spread. We position Nous as measuring the cognitive-monoculture problem and the limits of a prompt-level remedy, motivating deeper, below-the-prompt injection (fine-tuning, activation steering). Code, frozen profiles, prompts, and model outputs: https://github.com/WillChienT/nous-paper

Haowei Qian

Independent Researcher

Large Language Model Forecasting Cognitive Science Agent Behavior Modeling Frontier Models Prompt-Injection Detectors Extraction

Jun 10, 2026cs.AI

Rethinking Psychometric Evaluation of LLMs: When and Why Self-Reports Predict Behavior

Anticipating LLM behavioral tendencies from low-cost psychometric probes is critical for safe deployment, but only if self-reports (SR) reliably predict behavior. Recent work documented substantial SR-behavior dissociation in LLMs, but relied on broad personality traits (Big 5) that predict specific behaviors weakly, even in humans. Furthermore, the isolation of conversational sessions combined with weak context matching left open whether LLMs truly lack coherence or whether the conditions needed to detect such coherence were not met. We contrast Big 5 with the Theory of Planned Behavior (TPB), which measures intention targeted to a specific behavior and predicts human behavior substantially better than broad traits. We run experiments across four behavioral tasks and 11 frontier LLMs, while also varying session context and identity induction. We find that SR-behavior coherence exists but is selective. 1) Within a shared conversation, the Theory of Planned Behavior reaches human-level coherence; Big 5 does not. 2) Across separate conversations, coherence survives only for behaviors anchored outside the immediate prompt, such as implicit bias shaped by training, and collapses when behavior is strongly primed by context, as with sycophancy. 3) Persona prompting makes self-reports more consistent across conversations, but does not bring behavior into alignment. These findings suggest that coarse personality frameworks, such as Big 5 may not be the best tools for testing deployment behavior. More task- and behavior-specific instruments are needed, and even these must be evaluated across tasks and contexts.

Rafal Kocielnik, Pengrui Han, Peiyang Song +5

Caltech · UIUC · University of Cambridge

Personality Psychometric Properties Agent Behavior Modeling Self-Evaluation Social Reasoning

Jun 10, 2026cs.LG

Implicit Neural Representations of Individual Behavior

We study policy representation learning from unlabeled multi-policy behavioral data. Each episode is generated by a fixed policy, but policy labels are unavailable. This setting appears in robotics play, demonstrations, games, racing, and other datasets where heterogeneous behaviors are mixed without annotations. We introduce \emph{Behavioral INR}, a self-supervised generative model that adapts implicit neural representations (INRs) from vision to behavior. Instead of mapping coordinates to RGB values, Behavioral INR represents a policy as a state-action function mapping states to subsequent actions. An episode-level latent modulates this function through FiLM layers, yielding a generative prior over policies and allowing policy identity to be inferred without supervision. Because INRs treat each datapoint as samples from an underlying function, the same model naturally accommodates variable episode lengths and different sampling granularities, as in vision INRs with different image resolutions. We also define policy-level out-of-distribution (OOD) shifts along state-distribution and action-distribution axes, which arise when policies overlap in states or actions but are not captured by standard behavioral OOD settings based only on new agents or environments. We evaluate on synthetic Gaussian random field data, MuJoCo demonstrations with controlled OOD splits, and real-world chess, Formula 1 racing, robotics, and Seek-Avoid datasets. Behavioral INR most consistently improves policy identifiability in the hardest continuous state-action settings, especially when longer episodes, more policies, and OOD splits reduce the usefulness of marginal shortcuts; amortized history encoders remain competitive when policy identity can be recovered from symbolic repetition or low-dimensional action statistics. We release code and checkpoints.

Andrew Kang, Priya Narasimhan

Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, USA.

Neural Policies Neural Representations Agent Behavior Modeling

Jun 10, 2026cs.AI

SkillJuror: Measuring How Agent Skill Organization Changes Runtime Behavior

Agent Skills augment large language model (LLM) agents with procedural knowledge at inference time, but current benchmarks rarely distinguish what a Skill says from how it is organized. We study this distinction through Progressive Disclosure, where a concise root file points agents to supporting resources on demand, and compare it with a normalized flat baseline. We present SkillJuror, a framework for evaluating Skill writing paradigms through semantically controlled variants, matched multi-trial evaluations, and trajectory evidence while holding task knowledge fixed. In an 82-task SkillsBench study, Progressive Disclosure changes runtime behavior before aggregate outcomes: distinct Skill resources touched per trajectory rise from 1.18 to 3.85, and effective uptake events rise from 1.33 to 3.92. It also yields 17 additional verifier-passing trials out of 410 matched trials (+4.1%) over the normalized flat baseline. The benefit is task-dependent. Progressive Disclosure helps when supporting resources guide implementation, checking, or repair, but is weaker when success hinges on exact output conventions, numerical thresholds, or long artifact-generation pipelines. These results show that Skill organization is not mere presentation: it can change how agents search and apply procedural knowledge, while outcome gains depend on whether the exposed resources are actionable for the task. Code is available at https://github.com/zhiyuchen-ai/skill-juror.

Zhiyu Chen, Zihan Guo, Bo Huang +4

Tongji University · Shanghai Innovation Institute · Sun Yat-sen University +1

Agent Skill Retrieval Skills Disclosures Runtime Agent Behavior Modeling

Jun 9, 2026cs.RO

Steering Multirobot Behavior via Closed-Loop Affine Activation Editing

Real-world robots need to adapt their behavior beyond the envelope of their pre-trained policy. Policy finetuning or retraining are options, but they risk catastrophic forgetting, degrading the pretrained policy's base performance. To combat this, we introduce CLAE: Closed-Loop Affine Activation Editing, an inference-time framework for steering the behavior of a frozen policy by editing intermediate activations while keeping the base policy weights and downstream action head untouched. CLAE approaches behavior steering as a closed-loop problem whose outputs edit policy activations that adapt online to the robot state, environment, target behavior, and multi-robot context. It trains a sparse autoencoder over frozen-policy activations, selects behavior-relevant latent features via post-hoc probing, and learns a lightweight RL-based steering policy that applies state-dependent affine edits to selected latents during inference. We validate CLAE on a frozen multi-quadrotor navigation policy trained to perform a single task: navigating robots to a set of goal locations while avoiding obstacles. Through extensive simulations and physical tests, we show that while navigating to their goal positions, CLAE can 1. steer individual robot behavior by controlling each robot's velocity profile; 2. coordinate multirobot behavior by preserving a desired formation; and 3. produce entirely new behavior wherein robots are required to reduce their exposure to surveillance cameras in the environment. Our project website is available at https://clae-robotics.github.io/.

Satyajeet Das, Darren Chiu, Shashank Hegde +1

Department of Computer Science University of Southern California Los Angeles, CA, USA

Robot Policies Multi-Robot Systems Multi-Agent Reinforcement Learning Robot State Steering Quadrotor+2

Jun 9, 2026cs.AI

Forecasting Future Behavior as a Learning Task

Trust in an AI system is often anchored by explanations of how it works, which one then uses to forecast its behavior on new inputs. For large reasoning models (LRMs), this conventional route is particularly difficult to follow: explanation methods for single token generations do not naturally generalize to long trajectories, and the trajectories themselves are often not faithful when read as natural language. We propose an alternative that bypasses the explanation step: treat behavior forecasting as a learnable task and train Behavior Forecasters that operates on a single reasoning trajectory to make the same forecasts one would typically seek from an explanation. The forecaster's training data is obtained by querying the LRM with no human annotation, and its inference is done in a single forward pass. We instantiate this approach on two tasks: how likely the LRM is to repeat its answer on re-runs, and how removing parts of the input changes its answer. We evaluate this approach on both tasks across three diverse reasoning datasets and find that trained Behavior Forecasters are more accurate than GPT-5.4 and Claude Opus-4.6 reading the same trajectories as naive readers, at a small fraction of their inference cost. We find that fine-tuning the backbone end-to-end and initializing it from the target LRM are each necessary for strong performance. These results show that the reasoning trajectory carries information about the LRM's future behavior that goes beyond what naive reading conveys.

Mosh Levy, Yoav Goldberg, Asa Cooper Stickland

Bar-Ilan University · Constellation · UK AI Security Institute +1

Large Language Model Forecasting Reasoning Trajectory Large Reasoning Models Forecast Agent Behavior Modeling Explainable AI Methods+1

Jun 9, 2026cs.CL

ParaBridge: Bridging Paralinguistic Perception and Dialogue Behavior in Speech Language Models

Speech carries more information than just words: a child's voice, a fearful tone, or a noisy background should all lead a sufficiently competent spoken-dialogue assistant to different replies. Current Speech Language Models (SLMs) can recognize such paralinguistic cues but often ignore them in open-ended dialogue. We observe that a simple paralinguistic instruction scaffold at the inference stage narrows this perception-behavior gap, suggesting that the relevant cues are already latent in the model. Such scaffolds, however, remain brittle under multi-turn context and competing instructions. Therefore, we propose \textbf{ParaBridge}, an on-policy self-distillation method that turns a brittle inference-time scaffold into stable model behavior. During training, the scaffold serves only as a temporary privileged view; the scaffold-free model rolls out its own response, while the scaffolded view supplies dense, full-vocabulary next-token targets along its trajectory. This supervision teaches when non-lexical cues should affect the reply without the need for curated dialogues, human labels, or external reward models. On Qwen3-Omni-thinking, ParaBridge raises scaffold-free VoxSafeBench SAR from

14.6\%

to

40.3\%

and improves EchoMind average rating from

3.27

to

3.92

. It also preserves general ability, with MMAU-Pro, VoiceBench, and GPQA all within

0.4

points of the original model. Beyond the training distribution, ParaBridge generalizes to unseen paralinguistic cues, transfers from safety-oriented training to empathy-oriented dialogue, and works on a different SLM backbone.

Yuxiang Wang, Qinke Ni, Shengbo Cai +3

1The Chinese University of Hong Kong, Shenzhen · 2Tencent Hunyuan · 5Tsinghua University +2

Paralinguistic Cues Speech Language Models Dialogue Empathy Bridge Inference-Time+2

Jun 9, 2026cs.AI

Trace2Policy: From Expert Behavior Traces to Self-Evolving Decision Agents

Decision rules that enterprise experts apply tacitly -- in auditing, compliance, and contract review -- can be systematically recovered and improved through iterative error analysis. We present \textbf{Trace2Policy}, whose core mechanism -- \textbf{EISR} (\textbf{E}rror-driven \textbf{I}terative \textbf{S}kill \textbf{R}efinement) -- maintains a human-readable rule document as its optimization target: each round executes the rules on a validation set, clusters errors by root cause into MISSING, WRONG, or CONFLICT types, applies targeted patches, and commits only those that pass a regression gate. \textbf{For this class of compliance-sensitive, skewed-base-rate decision tasks, we identify rule quality -- not model capability -- as the dominant performance lever}: across five LLMs, one-shot distillation plateaus near

\sim

70% on the deployed pool, while eight EISR rounds lift the same rules to 79.6% when compiled into deterministic Python -- zero LLM calls at inference. \textbf{Execution form compounds the gain: in production, the same EISR-refined content runs 9.8~pp higher as compiled Python than as an LLM prompt, a form-and-engineering bundle the 22-day deployment matured together.} Deployed for 22 days at a major logistics carrier (3,349 audit cases), the compiled pipeline outperforms the pure-LLM baseline it replaced (72.7%); on these calibrated, skewed-base-rate workloads, re-enabling LLM fallback monotonically degrades accuracy. An LLM-driven variant, \textbf{Auto-EISR}, reproduces this refinement at $5--$10 per cycle versus

\sim

70 expert-hours, and transfers to four public benchmarks spanning legal reasoning (LegalBench) and process-mining decisions (BPIC 2012) without re-engineering.

Junli Zha, Jinbo Wang, Chao Zhou +1

SF Express

Large Language Model Policy Optimization Expert Demonstrations Agent Behavior Modeling Experts

Jun 9, 2026cs.AI

Mobility Anomaly Generation using LLM-Driven Behavior with Kinematic Constraints

Although the study of human trajectory anomalies is critical for advancing spatial data mining, empirical research remains severely hindered by a pervasive lack of ground-truth datasets. Despite the availability of several real-world and simulated human trajectory collections, these datasets exclusively capture normal mobility patterns and lack annotated anomalies. This specific scarcity is fundamentally driven by the inherent statistical rarity of anomalous events, precluding the feasibility of conventional observational methods. Compounding this challenge, the systematic acquisition of large-scale mobility data is strictly bottlenecked by prohibitive costs and stringent privacy regulations. To overcome these fundamental limitations and establish a reliable human trajectory anomalies dataset with annotated ground truth, we introduce a novel, end-to-end generative framework designed to synthesize realistic trajectory anomalies at scale. Our architecture bridges the gap between purely synthetic mobility data and complex real-world physical constraints by operating directly on baseline simulated trajectories. We employ Large Language Model (LLM) agents to systematically inject semantically meaningful behavioral anomalies such as irregular out-of-distribution check-ins and skipped routine visits. To ensure rigorous spatial validity, the system leverages map-constrained routing reconstruction to recalculate the physical transitions between these LLM agent-modified staypoints. Moreover, to narrow the simulation-to-reality gap, we augment the resulting trajectories with a context-aware spatial noise model, parameterized by environmental and location-specific variables, to accurately emulate heterogeneous GPS sensor degradation.

Yueyang Liu, Joon-Seok Kim, Andreas Züfle

Emory University, Atlanta, USA

Mobility Kinematics Sim-To-Real Gap Agent Behavior Modeling

Jun 8, 2026cs.CY

Interpretable Crisis Behavior Analysis Using Mobility and Social Media Data

Crises alter both how people move and how they communicate. During emergencies such as wildfires and pandemics, changes in mobility patterns and online emotional discourse evolve jointly, yet they are typically studied in isolation. This paper presents a unified and interpretable pipeline that integrates mobility and social media data to identify cross-domain behavioral patterns in crisis settings. The framework is evaluated through two case studies: a short-horizon analysis of the January 2025 Los Angeles wildfires (prototype case) and a longitudinal analysis of UAE COVID-19 behavior from March 2020 to December 2021 (primary case, 671 days). The pipeline aligns heterogeneous daily signals, transforms them into binary behavioral states, applies Formal Concept Analysis (FCA) to extract co-occurrence structure, mines association rules, and validates rule stability through chronological holdout testing. A structured policy-translation layer renders robust rules as operational briefs specifying triggers, lead times, and action playbooks. Results reveal clear cross-domain behavioral structure in both crises. In the wildfire case, traffic stress, fear/anger sentiment, and governance discourse are tightly coupled within a 33-day window, with key rules reaching 100% confidence and lift scores up to 2.5. In the COVID case, repeated mobility adaptation and sentiment volatility yield 8 stable same-day rules (88% holdout pass rate) and 40 clean predictive rules with 2--7 day lead horizons. The work demonstrates that interpretable multimodal fusion can produce both scientifically credible and policy-actionable crisis intelligence.

Muhammad Hamza Arshad Majeed, Sidahmed Benabderrahmane, Talal Rahwan

New York University (NYUAD), Division of Science, Computer Science Department

Crisis Covid-19 Mobility Agent Behavior Modeling

Jun 7, 2026cs.RO

Guided Discovery of New Behaviors using Diffusion Policies

Diffusion models have become a powerful tool for generative modeling in robotics, with diffusion policies excelling at modeling multimodal action-trajectory distributions. However, when demonstrations are limited, standard sampling often reproduces dominant behaviors while neglecting valid but rare modes, limiting the discovery of novel solutions. Existing approaches, such as guidance methods or combining reinforcement learning with diffusion, either push samples into infeasible regions or struggle to escape local minima, failing to systematically uncover diverse behaviors. To address these challenges, we propose a framework that combines Feynman-Kac correctors with a novel guiding potential that systematically guides diffusion policy samples towards promising yet underrepresented samples. These trajectories are refined using sampling-based trajectory optimization and reincorporated into the training set to retrain the diffusion policy. Our method effectively mines and repairs novel trajectories, enabling the systematic discovery of diverse and executable behaviors. We demonstrate the effectiveness of our framework across a range of manipulation environments, consistently discovering new behaviors.

Dian Yu, Sebastian Sanokowski, Majid Khadiv

Munich Institute of Robotics and Machine Intelligence, Technical University of Munich

Diffusion Policies Robot Policies Multimodal Action Distributions Agent Behavior Modeling Generative Models

Jun 7, 2026cs.CL

More Yap Less Meaning: Uncovering Self-Improvement Behavior in SLMs

Recently, language models have made rapid progress across various domains and applications. However, their capability for self-improvement, i.e., whether they are adept at recognising and correcting flaws in their own reasoning, remains dubious. In this study, we address this question by constructing a sufficiency test to rigorously examine the self-correction capabilities of small language models (SLMs). We propose a minimal three-step self-correction pipeline that collects initial SLM answers, prompts the same model to generate hints for its incorrect responses given the ground truth, and feeds the model the same question with its own feedback to refine the initial answer. We evaluate a variety of instruction-tuned and reasoning SLMs in this experimental setup on arithmetic and logical reasoning benchmarks. Our findings show that SLMs with injected hint sentences yield only a 4.4 percent gain over initial question-answering accuracy. Even though the correct answer was provided alongside the model's incorrect reasoning, the evaluated SLMs fail to understand what was missing in their reasoning and show minimal semantic difference between hints that lead to corrections and ones that do not. Furthermore, our experiments show that longer hints are positively correlated with incorrect final answers, suggesting that longer deliberation on problems can hinder the reasoning process, meaning that SLMs do not necessarily scale in performance with a larger compute budget.

Marina Igitkhanian, Erik Arakelyan

American University of Armenia · NVIDIA

Small Large Language Models LLM Reasoning Strategies Recursive Self-Improvement Large Language Model Responses Reasoning Benchmark Agent Behavior Modeling

Jun 7, 2026cs.LG

Beyond Linear Activation Steering: Invertible Latent Transformations for Controlling LLM Behavior

Activation steering provides a lightweight inference-time mechanism for controlling large language models (LLMs) by modifying their internal activation vectors toward desired behaviors. Most existing methods compute a fixed steering direction in the original activation space, typically from pairs of contrastive examples using mean differences, linear probes, or arbitrary separability criteria. While effective to a certain extent, these methods treat behavioral control as a global, linear, additive offset: the same direction is applied across inputs, and behaviors are linearly separable. This can be restrictive when behavioral features vary nonlinearly across the activation space or lie on curved and anisotropic manifolds, where the optimal intervention may be input-dependent. To address this limitation, we propose INNSteer, a nonlinear activation steering framework based on invertible latent transformations. Rather than searching for a better steering vector in the original representation space, INNSteer learns a lightweight invertible neural network

φ

that maps an LLM's activations into a latent space where behavioral classes are more amenable to linear control. At inference time, activations are mapped through

φ

, steered in the latent space, and mapped back through the exact inverse transformation

φ^{-1}

. This makes a simple latent-space translation become a nonlinear, input-dependent intervention in the original activation space. Across experiment settings on multiple LLM families, scales, behavioral traits, and safety benchmarks, INNSteer consistently improves model control over linear, transport-based, and nonlinear steering baselines while largely preserving generation fluency.

Tuc Nguyen, Thai Le

Indiana University Bloomington, USA

Linear Activation Steering Steering Model Activations Latent Feature Interventions Agent Behavior Modeling

Jun 5, 2026cs.CL

Representational Similarity and Model Behavior in Multi-Agent Interaction

Researchers have shown that neural similarity among humans predicts social closeness and cooperative success, whereas innovation often emerges from interactions among dissimilar individuals. We investigate whether these principles extend to artificial intelligence by examining interactions between large language models. In our experiments, 276 model pairs interact across eight games spanning both cooperation and novelty. We find that pairs with more similar representation spaces achieve significantly higher cooperation but exhibit reduced novelty and creativity. The effects of representational similarity on cooperation and novelty remain robust even after controlling for other factors such as performance disparity and model size. We also find that similarity in the early layers consistently shows the strongest association with cooperation and novelty, compared to the middle and later layers. This suggests that a central factor underlying these patterns could be the extent to which the two models share lexical and semantic grounding. Overall, representational similarity can be an important consideration in multi-agent system design.

Yujin Potter, Seun Eisape, Shiyang Lai +6

University of California, Berkeley · The University of Chicago, Knowledge Lab · Santa Fe Institute +2

Cooperation Similarity and Metrics Agent Behavior Modeling Proximity Operations

Jun 5, 2026cs.CV

Automated 3D Kinematic Monitoring for Circadian Activity and Anomaly Detection in Juvenile Fish

Precision aquaculture faces a "phenotyping bottleneck" in tracking high-resolution behavioral traits, as conventional methods cannot quantify instantaneous three-dimensional (3D) physical exertion. To address this, we present a high-throughput 3D behavioral phenotyping framework integrating deep learning object detection with binocular stereo vision for real-time monitoring of juvenile tilapia in high-density environments. The system automates non-contact body length estimation and reconstructs 3D swimming trajectories from absolute spatial coordinates. By eliminating 2D perspective distortions, this approach precisely quantifies 3D velocity and acceleration, marking the first estimation of true physical swimming speeds in free-roaming juveniles. Results show the framework successfully establishes circadian locomotor baselines, serving as an early warning system for physiological stress and providing an objective metric for fish vitality.

Chih-Wei Huang, Chang-Wen Huang, Chung-Ping Chiang +1

AI Research Center, National Taiwan Ocean Univ., Keelung City 20224, Taiwan. · Dept. of Aquaculture, National Taiwan Ocean Univ., Keelung City 20224, Taiwan. · Center of Excellence for the Oceans, National Taiwan Ocean University, Keelung City 20224, Taiwan.

Fish 3D Generation Agent Behavior Modeling

Jun 4, 2026cs.AI

Entropy-Based Observability for AI Agent Behavior

AI agents are typically instrumented through outcome-oriented indicators such as task success, reward, latency, and cost.Although these indicators are operationally important, they provide limited visibility into the internal structure of agent behavior such as the degree of exploration, the rigidity or diversity of action selection, the concentration of tool use, the reduction of uncertainty across a run, and the stability of behavior across repeated executions.This paper proposes Entropy-Based Observability for AI Agents (EOA), a lightweight framework for deriving behavioral telemetry from agent traces.

Olasimbo Ayodeji Arigbabu

Artificial Intelligence Agents Partial Observability Agent Behavior Modeling Action Selection

Jun 2, 2026cs.CY

Agentic Safety is an Epistemic Property, Not a Behavioral One

Contemporary AI safety spans pre-training interventions, post-training alignment, deployment-time controls, monitoring, and red-teaming. These methods are necessary, but they primarily certify snapshots of system behavior. As AI systems become more capable, dynamic, embodied, and self-improving, this snapshot view becomes incomplete: safety depends not only on whether a system behaves acceptably now, but whether it remains correctable as it learns, adapts, acts, and modifies itself over time. This paper argues that safety should therefore be treated as an epistemic property of the evolving learner, not merely a behavioral property of the current policy. We introduce teachability as the capacity to preserve future corrective leverage under bounded human, institutional, or environmental intervention. We argue that advanced systems can retain visible competence while eroding the representational, algorithmic, or meta-decision conditions needed for future correction. Safe advanced AI systems must not only behave acceptably now; they must remain teachable later.

Charles L. Wang, Keir Dorchen, Peter Jin

1Columbia University, New York, NY, USA · 2Pelagic Platforms

Artificial Intelligence Safety Epistemic Integrity Agent Behavior Modeling Decisions

Jun 2, 2026cs.AI

Decomposing how prompting steers behavior

Prompting steers large language models (LLMs) and vision-language models (VLMs) without weight updates, but it remains unclear how instruction changes reshape internal representations to produce behavior. We introduce a nested geometric decomposition framework that treats prompting as a transformation of the representational geometry of the content following the prompt. For each prompt pair, we align representations of the same stimuli under two prompts using increasingly expressive stimulus-invariant maps: translation, rigid transformation with uniform scaling, sequential axis scaling, affine transformation, and nonlinear transformation. We then causally test each map by replacing a single layer's prompt-A hidden state for held-out stimuli with its mapped counterpart and measuring recovery of prompt-B representational geometry and behavior. Across three LLMs, three VLMs, and six text or image datasets spanning style, emotion, scene content, and number, prompts consistently reshape representations toward the instructed task structure. Cross-validated variance decomposition shows that much prompt-induced activation change is captured by shape-preserving maps, especially translation and rigid transformation with uniform scaling, while tier profiles reveal model- and task-specific routing strategies across layers. Crucially, although translation and rigid tiers already improve behavioral agreement, affine transformation is the first tier to nearly recover target-prompt task geometry and yields corresponding behavioral gains. This suggests that cross-dimensional linear mixing is a key mechanism by which prompts reorganize representations toward instructed task structure. Our framework decomposes prompt-induced representational change into interpretable geometric components and reveals how models route task-relevant structure to produce prompt-driven behavior.

Fan L. Cheng, Nikolaus Kriegeskorte

Columbia University

Instruction-Tuned Models Steering Agent Behavior Modeling Geometric Constraints

Jun 1, 2026q-bio.PE

Self-Regulation through Communication in Evolved Neural Agents

Communication is typically understood as indication: signals that transfer information from sender to receiver. We present a minimal predator avoidance task in which pairs of evolved CTRNN agents use communication for robust survival, and in which agents hear their own vocalizations, as in natural systems. Across 112 perfect-fitness agents from over 2,000 evolutionary runs, three dominant strategies emerge (accounting for 81% of agents): safety calling (39%), where agents signal from safe cover; alarm indication (22%), where agents vocalize when a threat is present without relying on self-hearing; and self-regulatory calling (20%), where agents depend on hearing their own call to sustain escape behavior. Self-hearing dependency is common among agents that call during an active threat (47%), but rare among agents that call only after reaching safe cover (10%; p < 10^-4). This pattern is consistent with a difference in causal order: safety callers act then communicate, while self-regulatory callers communicate in order to act. Removing self-hearing selectively impairs self-regulatory callers (fitness 0.40) while safety callers remain functional (0.90; p < 10^-9). These results show that communication can evolve to serve the caller's own behavioral regulation, not just information transfer to others.

Joshua Nunley

Indiana University, Bloomington, IN, USA

Predator-Prey Models Self-Evolving Agents Agent Behavior Modeling

Jun 1, 2026cs.AI

BehaviorBench: Modeling Real-World User Decisions from Behavioral Traces

Many decision-support settings require systems that adapt to individual users, but evaluation data for this problem remain limited. Existing benchmarks for user understanding often rely on simulated users or model-generated behavior, even though recent work cautions that model-based simulations can diverge systematically from human behavior. We introduce \textsc{BehaviorBench}, a benchmark for evaluating personalized decision modeling from real-world behavioral traces. \textsc{BehaviorBench} reconstructs wallet-level decision histories from observed public prediction-market and on-chain records, and organizes them into two complementary task layers: \emph{Belief prediction}, which predicts a user's final revealed stance and confidence in a market, and \emph{Trade prediction}, which predicts the direction and amount of individual transactions. Across 2,000 evaluation wallets, the benchmark contains 141,445 Belief instances and 1,485,972 Trade instances, with disjoint support pools for retrieval-based evaluation. We evaluate frontier and open-weight generative models under four history interfaces: no personalization, direct recent history, generated user profiles, and retrieved support-wallet evidence. Personalization improves Belief prediction more consistently than Trade prediction, model rankings change across task layers and metrics, and different history interfaces expose different failure modes. \textsc{BehaviorBench} provides an evaluation setting for studying whether personalized methods can use real-world behavioral evidence rather than simulated users alone.

Liangwei Yang, Jielin Qiu, Zixiang Chen +9

Salesforce AI Research

User Modeling Behavioral Foundation Models Depth History Agent Behavior Modeling User Simulation

Jun 1, 2026cs.CL

Greener Than Humans? Environmental Attitudes in Large Language Models

Large language models (LLMs) are increasingly used in sustainability-related decision support, reporting, and public communication, yet little systematic evidence exists on the environmental attitudes embedded in their outputs. This paper develops a benchmark for evaluating environmental cognition, affect, and behavioural recommendations in LLMs and applies it to 31 widely used proprietary and open-weight models. Drawing on questions from established environmental awareness surveys and additional sustainability-related behavioural measures, we compare LLM responses 1) among models and 2) between models and human survey benchmarks from Germany. We assess their robustness across prompting conditions. We find that many LLMs align more closely with environmentally progressive attitudes than the average survey respondent, exhibiting higher levels of environmental affect and cognition and recommending behaviours associated with substantial potential CO2 reductions. At the same time, we observe no systematic relationship between sustainability-oriented responses and model origin, size, or release context. However, models exhibit contextual sensitivity, controlled by persona-based prompting and show sycophantic shifts mirroring user-specified ideological positions, which raises concerns about steerability and normative reliability in real-world deployments. Our findings provide a reusable evaluation framework for assessing sustainability-related value alignment in LLMs and highlight the importance of governance, transparency, and critical oversight as AI systems become increasingly embedded in sustainability transformations and public decision-making.

Stefanie Kunkel, Tilman Hartwig, Marcus Voss +2

Research Institute for Sustainability at GFZ Centre for Geosciences, Berliner Straße 130, 14467 Potsdam, Germany · German Environment Agency, Wörlitzer Platz 1, 06844 Dessau-Roßlau, Germany · Birds on Mars, Marienstraße 10, 10117 Berlin, Germany

Sustainability Environment Agent Behavior Modeling

Jun 1, 2026cs.AI

Tracking the Behavioral Trajectories of Adapting Agents

Text files such as skill files, memory files, and behavioral configuration files play a central role in defining how modern agents act. Through edits by humans or the agents themselves, these files may evolve over time, directly steering the agent's behavior in future interactions. We present a methodology and framework for measuring agent

traits

by defining traits as directions in the embedding space of a text embedding model. We train a linear model on labeled "before" versus "after" skill file diffs to learn a trait vector, then score arbitrary skill edits by projecting their embedding diffs onto this vector. Evaluated on 68 labeled skill diff pairs for the trait of propensity to seek sensitive data, our method achieves 91.2% sign classification accuracy and a Spearman rank correlation of

ρ= 0.82

under leave-one-out cross-validation. We build this trait evaluation into a broader agent-to-agent protocol that enables one agent to evaluate another's skill file updates through a trusted intermediary.

Jonah Leshin, Manish Shah, Ian Timmis

Project VAIL, San Francisco, CA, USA

Agentic Evaluations Skills Agent Behavior Modeling

Jun 1, 2026cs.LG

Coherent Off-Policy Improvement of Large Behavior Models with Learned Rewards

Distilling expert demonstration data into large generative models using behavioral cloning is a scalable approach to learning capable policies for robotic control, particularly for dexterous manipulation. Reinforcement learning (RL) can be used as a means to finetune these policies further using additional experience. An open question is whether RL is more sample-efficient than collecting more human demonstrations. Prior work has finetuned large pretrained policies in a scalable fashion by applying RL to a smaller residual policy that corrects the pretrained model. However, for the typical sparse reward tasks, RL algorithms can struggle to optimize the behavior in a sample-efficient manner. We explore inverse reinforcement learning, where a dense reward function is learned from expert demonstrations, potentially reducing the challenge of RL finetuning. We specifically consider coherent imitation learning, an IRL method that facilitates improvement of the BC policy through using a specific reward formulation with theoretical guarantees. We show that our IRL method maintains or improves the performance of pi-0.5 on all six sparse manipulation tasks and achieves a

\geq 90\%

success rate on five out of six complex manipulation tasks, outperforming RL-based baselines using sparse rewards. By ensuring our initial pretrained finetuning policy is optimal for our initial reward and critic, our method circumvents the initial drop commonly seen in RL finetuning and enables faster improvement.

Christian Scherer, Joe Watson, Theo Gruner +3

Technical University of Darmstadt · Zuse School ELIZA · University of Oxford +2

Reinforcement Fine-Tuning Dexterous Manipulation Agent Behavior Modeling Improvement Model Size

Jun 1, 2026cs.CL

CRAB-Bench: Evaluating LLM Agents under Complex Task Dependencies and Human-aligned User Simulation

Evaluating LLM agents in realistic service scenarios requires complex task dependencies, imperfect user behavior, and an evaluation that accommodates multiple valid solutions. We introduce CRAB-Bench (Constraint-based Realistic Agent Benchmark) and RUSE (Realistic User Simulation Engine) to address this gap. CRAB-Bench generates tasks via a constraint graph over multiple interdependent entities with structured distractors, requiring agents to reason carefully over thousands of misleading candidates where only a tiny fraction of solutions are valid. RUSE replaces cooperative, template-like simulators with realistic users grounded in human behavioral studies, instantiated across diverse personas and four behavioral dimensions. Experiments on four frontier LLM agents show that the best model achieves only 61% pass@1 on CRAB-Bench, and switching to RUSE causes further drops of up to 57%, concentrated in task-solving ability rather than conversational quality. Information Disclosure is the most damaging behavioral dimension, and agents interacting with RUSE are less likely to admit mistakes, instead masking errors through implicit corrections.

Danqing Wang, Akshay Sivaraman, Lei Li

Carnegie Mellon University

User Simulation Agentic Benchmarks Agent Behavior Modeling

May 31, 2026cs.CL

Understanding LLM Behavior in Multi-Target Cross-Lingual Summarization

Multi-target cross-lingual text summarization (MTXLS), which summarizes a source document into multiple target languages, is increasingly important as users consume content in diverse languages, but remains underexplored. To address this gap, we introduce multi-target cross-lingual element-aware (MEA), a new MTXLS benchmark covering 24 target languages. We benchmark end-to-end and pipeline approaches across various LLMs and show that MTXLS performance still substantially lags behind English monolingual summarization. To better understand MTXLS in LLMs, we propose a layer-wise analysis framework for investigating how LLMs internally perform MTXLS. Our analyses suggest that translation and summarization behaviors emerge jointly within later layers rather than as distinctly decomposed stages. Most task-relevant processing occurs within these layers, and errors also tend to arise at similar depths. Motivated by these findings, we introduce an inference-time activation steering method that leverages hidden representations from English summarization to guide MTXLS generation. Experiments show that our method consistently improves MTXLS quality across target languages.

Sangwon Ryu, Yihong Liu, Mingyang Wang +4

GSAI, POSTECH · LMU Munich · MCML +2

Cross-Lingual Transfer Summarization Spreadsheets Agent Behavior Modeling

May 31, 2026cs.AI

Diagnosing LLM Arbitration Behavior over Pre-evidence Epistemic States in RAG-based Fact-Checking

In RAG-based fact-checking, LLMs are increasingly used as verifiers to check given claims against retrieved evidence. Their parametric knowledge can induce pre-evidence tendencies that may conflict with the retrieved context, yet existing evaluation frameworks do not characterize such prior-context discrepancy or measure how verifiers arbitrate between parametric and contextual signals. We introduce \textsc{PAVE} (\emph{Prior-Aware Verifier Evaluation}), a diagnostic testbed that stratifies an LLM verifier into four epistemic states based on the correctness and confidence of its pre-evidence prior and evaluates its arbitration behavior on this new benchmark, i.e., whether it persists in correct prior under misleading evidence, and whether it corrects wrong prior when accurate evidence is provided. Experiments across seven LLMs reveal unreliable and highly model-dependent prior-context arbitration, highlighting the importance of verifier selection for real-world RAG-based fact-checking applications. Based on these findings, we propose a lightweight JSD-based test-time arbitration method that improves factual reliability without modifying the underlying model, achieving competitive performance across diverse LLM families.

Yuxi Sun, Wenbo Shang, Wei Gao +2

Hong Kong Baptist University · Singapore Management University

Fact-Checking Large Language Model Reliability Epistemic States Agent Behavior Modeling

May 31, 2026cs.RO

Position: Good Embodied Reward Models Need Bad Behavior Data

This position paper argues that to obtain reliable embodied reward models, the community must invest in ``bad'' robot data: failed, suboptimal, error-prone, and even hazardous behaviors. While reward models are central to any foundation model's lifecycle, today's embodied reward models are trained primarily on successful behaviors. We analyze three state-of-the-art embodied reward models and find that they systematically over-reward behaviors that real human evaluators would penalize, including unsafe interactions, poor execution, and shortcut strategies that only superficially satisfy tasks. We attribute these failures to a key data gap: the scarcity of negative embodied data which is costly to collect and often filtered out or withheld in existing robotics datasets. Furthermore, we show that even modest exposure to real bad behavior data can improve alignment with human preferences and reduce costly false positives. We therefore call on the embodied AI community to curate and release their bad robot data, build synthetic bad data generation engines, develop more decentralized physical evaluation systems, and design benchmarks for fine-grained embodied reward model evaluations.

Ran Tian, Yilin Wu, Andrea Bajcsy

UC Berkeley · NVIDIA Research · Carnegie Mellon University

Robotic Data Acquisition Agent Behavior Modeling Bad

May 31, 2026cs.AI

Prospect-Theory Behavior from Bellman Optimality in MDPs with Catastrophic States

We study risk-neutral control in Markov decision processes with an absorbing catastrophic state. Even though rewards are linear and the agent has no utility curvature, probability weighting, or framing dependence, standard Bellman optimality produces three prospect-theory-like signatures: an S-shaped value-function profile (convex near catastrophe, concave in the far field), an endogenous loss-sensitivity coefficient

λ^*(S) > 1

, and a reflection-effect policy reversal. Across 495 configurations, the optimal policy plays safe near catastrophe in positive-drift (growth) regimes despite the risky action's higher immediate expected value, and plays risky near catastrophe in negative-drift (decline) regimes despite the safe action's lower immediate expected loss. We derive a closed-form expression for the asymptotic loss-aversion plateau

\barλ

that depends only on win probability

p

, payoff asymmetry

r = |Δ_\ell/Δ_w|

, and discount factor

β

, and matches numerical solutions to

R^2 = 0.999

. The mechanism does not require asymmetric payoffs. Across a sweep of

(p,β)

at three asymmetry levels, the asymmetry share of

\barλ

above unity has median 4.6% at

r = 1.25

and rises to 13.9% at

r = 2

, with the boundary contribution exceeding the asymmetry contribution in every cell tested. The phenomena persist under tabular Q-learning (a model-free agent reproduces

V^*

at correlation 0.98 in growth and 1.00 in decline) and under stochastic transitions with Gaussian, heavy-tailed Student-

t_3

, and asymmetric skew-normal noise up to 50% of the step size, where the asymptotic plateau tracks the closed-form prediction within 0.41% for safe-channel noise and within 9.6% for risky-channel or both-channel noise. These results identify absorbing failure states as a sufficient structural mechanism for prospect-theory-like behavior under optimal control.

Yujiao Chen

Massachusetts Institute of Technology · Cambridge, MA 02139, USA

Conditional-Value-At-Risk Markov Decision Processes Bellman Equation Agent Behavior Modeling Distributional Signatures