Role

Informational active matter shows how measurement-informed decisions produce collective order, so far in systems that reach consensus. We design collective information engines structured by differentiation instead, and construct a minimal instance using anti-coordination games where differentiated role information has value. Within many coexisting games, agents infer their role from a noisy social signal grounded in a persistent identity, and role-following action feeds back into that signal, which shapes the incentive to follow roles. Resources accrued through coordinated role-play combine with identity variability to reinforce the schemas that generated them. The model thereby operationalizes Sewell's duality of schemas and resources in Structuration, a resolution to structure--agency debates across social science. The engine ignites when a social loop gain---the product of identity persistence, cognitive capacity, channel fidelity, and schema strength---exceeds one. For a repertoire of such schemas, roles emerge with increasing gain in a bifurcation cascade whose functional form is fixed by the repertoire's eigenvalue spectrum, ranging from monitorable logarithmic sequences to avalanches that arrive without warning. Resource accumulation supplies the fitness of a replicator dynamics on schema strengths, which selects the cascade type endogenously. Subcritical identity covariance reveals that type before onset, enabling early detection, while feedback channel parameters bias which type is selected. Platform design then becomes a control lever to throttle emergent coordination. This theory grounds distributional AGI takeoff in a mechanism and provides a monitor-based solution. Joining game theory, collective dynamics, and information engines, we open a route to an information thermodynamics of agent populations.

Sustained deployment of generative AI agents requires more than isolated task success. Agents must remain useful across repeated interactions, changing conditions, and dependencies on people within shared workflows, especially as technical, human, and operational disruptions accumulate over time. We propose operational resilience and considerate participation as two complementary aspects of evaluating such agents: the former captures how agents recover from blocked work while preserving progress and communicating their limits, and the latter captures how their adaptation accounts for affected people, role boundaries, and the surrounding workflow. Yet both remain underexplored under accumulating challenge. We study 120 simulated healthcare trajectories across two generative AI models and twelve stakeholder-derived tasks under light, medium, and heavy challenge. We compare textual action plans, prompted internal assessments, and quantitative structured workload and affect reports to examine how agent behavior and reported state change as challenge accumulates. Regarding operational resilience, agents shift from self-directed recovery toward greater human dependence, while reporting increasing workload and negative affect in structured reports but seldom expressing strain in textual responses. Regarding considerate participation, agents broaden from task-focused adaptation toward task reframing, attention to others, role-boundary adjustment, and wider coordination, with distinct patterns across actions and internal assessments. From these findings, we derive five deployment dilemmas involving persistence, attention, role boundaries, state disclosure, and escalation that require stakeholder specification, further informing technical implications for learning, situated evaluation, and embodied adaptation.

Large Language Models (LLMs) often exhibit "Attention Sink" (AS) and the accompanying "Massive Activations" (MAs) at the initial position of a sequence. These phenomena frequently co-occur, and MAs can pose challenges for low-bit quantization. In this study, we analyze the factors underlying AS and MAs that emerge at the initial position regardless of the token occupying it. Our experiments suggest that self-concentration of attention, resulting from the causal mask, and the subsequent Value-non-mixing in attention outputs contribute to AS and MAs. These findings provide new empirical evidence on the internal dynamics of LLMs, offering insights that may inform future quantization strategies and advance our understanding of the internal mechanisms of attention layers.

Large Language Models (LLMs) are increasingly being investigated for physiological time-series prediction, yet their effectiveness may depend not only on the model itself, but also on how physiological information is represented and presented at inference time. This study investigates prompt-based general-purpose LLMs for postprandial hyperglycemia and hypoglycemia prediction in individuals with type 1 diabetes. Using the OhioT1DM dataset, we evaluate multiple open-weight LLMs under zero-shot and few-shot inference across prediction horizons of 30, 60, and 90 minutes. The analysis varies both the textual representation of the available physiological information and the amount of information exposed to the model, ranging from glucose observations alone to derived descriptors and additional contextual variables related to insulin, meals, carbohydrates, and physical activity. Performance is compared with conventional patient-specific supervised models and with Gluco-LLM, a language-model-based architecture explicitly adapted to glucose time-series forecasting. Results show a marked task-dependent behavior. Conventional supervised models achieve the strongest performance for hyperglycemia prediction, whereas the best observed prompt-based LLM configurations improve performance for hypoglycemia across all investigated horizons. The effectiveness of prompt-based inference is also strongly influenced by how physiological information is represented, while providing additional contextual information does not lead to a systematic improvement. Overall, these findings highlight physiological information representation as a central design factor in prompt-based LLM approaches to glycemic-event prediction.

LLM-based user simulators are increasingly used to evaluate autonomous agents at scale, in place of costly human evaluations. Despite this promise, these simulators exhibit "assistant bias," a tendency to cooperate and pursue task goals. They rarely reproduce the frustration or disengagement that real users exhibit, compromising evaluation validity. Prior work outlines that this bias is baked in during model training, which role-playing prompts fail to override. We analyze this bias from model activations, extracting a user role vector by contrasting how the model represents user versus assistant perspectives on the same dialogue. We observe two findings: (i) the user direction is identifiable in activations, elicits user-like behaviors, and captures characteristics distinct from assistant traits; and (ii) although user-role activation associates with simulation realism and steering strengthens it, it can exaggerate user behaviors and override individual user profiles. Together, our findings provide a representation-level analysis of LLM user simulators, confirming that assistant bias is structurally identifiable and that user behavior can be directionally analyzed.

Vision-Language-Action (VLA) models are reshaping autonomous driving (AD) by unifying perception, reasoning, and control through language, enabling semantic grounding, interpretable decisions, and better long-tail generalization. But language is expensive onboard: latency and memory budgets are tight, and autoregressive decoding is inherently sequential. This work reframes the central question as when and where language should act at inference, since inference cost recurs at every deployed frame while training cost is paid once. We introduce the Language Residue taxonomy to organize methods by their inference-time use of language: train-time-only supervision (L1), latent non-textual reasoning (L2), conditional invocation (L3), and full per-frame generation (L4). We review representative methods and tag each across five deployment axes (latency, parameters, memory, FLOPs, tokens), analyzing them on major open- and closed-loop driving benchmarks (e.g., nuScenes, NAVSIM, Bench2Drive). We further trace how efficient methods from NLP/LLM are adapted in AD, identifying the constraints and motivations driving these adaptations. A continuously updated repository will be available at Github.

Roles provide an interpretable interface for organizing language-model agents, yet most multi-agent systems treat them as hand-written prompt labels disconnected from learned behavior and parameter updates. We argue that a useful role should instead be an executable control variable: it should summarize behavior predictive of future utility, guide subsequent interaction, and identify the trainable capacity responsible for that behavior. We introduce ExRole, a trajectory-to-role framework that learns future-aware role prototypes from prefix-local team traces, resolves them into readable instructions and token-aligned role markers, and optionally routes shared LoRA rank slots with turn-aligned credit. Across MuSiQue and 2WikiMultiHopQA, ExRole improves over single-agent search by 15.0/14.4 and 13.5/16.1 EM/F1 points, respectively. Against the strongest non-ExRole controls, the corresponding gains remain 11.5/11.6 and 7.7/9.7 points. Across both benchmarks, the controlled results consistently favor trajectory-induced role conditioning over role-free, manual, random, and shuffled alternatives. Role-Agent-Turn interventions further show that the induced roles capture transferable behavioral specialization beyond fixed agent identities or turn positions.

Soft skills shape collaboration among ML engineers, data scientists, and software engineers building ML-enabled systems, yet what we know about them comes almost entirely from the demand side. Job advertisements, surveys, and hiring manager interviews capture what employers ask for. How candidates themselves articulate these competencies has not been studied, and existing CV-mining work is both keyword-based, so it cannot see skills conveyed through narrative, and descriptive, reporting frequency rankings without testing whether group differences exceed sampling variation. We close both gaps. Using a balanced corpus of 300 curated CVs spanning the three roles, we extract explicitly listed and implicitly narrated soft skills with an LLM-based pipeline validated against a human-annotated ground truth, a distinction that existing extractors were not designed to make. We then convert the demand-side literature's claims into 13 falsifiable hypotheses about role signatures, seniority progression, and disclosure style, and test them with effect sizes under family-wise error control, so that candidate-side data can corroborate or contradict the demand-side account rather than merely illustrate it. Eleven hypotheses are supported, one partially, and one refuted. Candidates disclose soft skills through narrative rather than keyword lists by roughly three to one, and most so for the competencies employers value most: leadership, coordination, and mentoring (88-96% narrative). Seniority nearly triples the odds of articulating leadership. That competency, assumed universal in prior work, is articulated by software engineers at half the rate of their peers. Technical candidates do articulate soft skills, but a keyword-based screening systematically misses them.

Large multi-call LLM systems pose a scientific problem that query-level routing does not capture: the value of a model depends on where it enters a dependent computation and on the deployment that surrounds that call. Existing routers typically decide \emph{where} to spend a stronger model while treating the benefit of the substitution itself as known. We separate these two decisions through a predicate-action factorization and evaluate it in controlled solve-merge-verify workflows spanning 8-64 solve calls and four three-tier model ladders. The resulting evidence reveals a consistent principle beneath apparently conflicting outcomes. On numeric frequency counting, all-strong reduces RMSE from 4.818 to 1.538 in the Mixed Qwen/GPT ladder, whereas the average Qwen-only ordering reverses. Input-matched interventions further show that the same medium-to-strong action has sharply different value across roles and scales. A semantic task-and-contract shift reverses the Mixed ordering again, while allocation ablations distinguish useful sparse placement from under-coverage and indiscriminate escalation. Together, these results establish model substitution as a deployment-conditioned action rather than a property implied by a tier label, and they provide a practical sequence for large-scale workflow routing: calibrate the action, resolve its role-conditioned effect, and then optimize its placement.

Negotiation is a demanding social task for LLM agents, requiring strategic reasoning, persuasion, and interpersonal adaptation. Yet existing benchmarks often treat agents as emotionally neutral, overlooking a key driver of human bargaining behavior. We study how prompt-conditioned emotions affect LLM-based price negotiation. In a controlled framework, buyer and seller agents are independently assigned one of six emotional states and negotiate over 350 real consumer products under two budget conditions. Across 36 emotion-pair settings and five widely used LLMs, we find that emotions strongly shape outcomes. Angry buyers almost never reach agreement (0.39% deal rate), while happy buyers agree most often (28.91%), but obtain worse prices than fearful buyers. Emotion effects are role-dependent: buyer emotion mainly drives acceptance and rejection, whereas seller emotion shapes concession dynamics. These effects influence not only language, but also termination behavior and price trajectories, raising concerns for emotion-conditioned agents in commerce.

Mixture-of-experts vision-language models (MoE-VLMs) increase model capacity with sparse expert activation, yet deployment requires storing the full expert pool. Training-free expert merging reduces this burden, and many routing-based methods aggregate routing statistics across all tokens to determine merge compatibility. However, MoE-VLM inference is phase-structured: image-context tokens carry visual content, question tokens specify the query, and answer tokens produce the output, with different counts and routing distributions. Because image-context tokens are far more numerous, global aggregation can overemphasize image-context processing and obscure phase-conditioned expert roles, making experts serving different phases appear interchangeable and degrading model performance. We therefore argue that MoE-VLM expert merging should preserve phase-conditioned expert roles, judging compatibility by how experts serve different phases rather than globally aggregated routing statistics. Based on this view, we propose RoleMerge, a training-free method that constructs each expert's Routing Role Profile (RRP) from phase-normalized routing statistics, capturing its relative phase preference. Guided by expert-phase information loss, RoleMerge merges experts with compatible profiles and their corresponding router entries while preserving answer-decoding expert distinctions. Experiments on three models and multiple benchmarks show that RoleMerge preserves more of the full model's performance than alternative expert-merging methods at matched expert-retention ratios, with relative improvements of up to 9.6 percent in six-task macro-average performance. These results validate phase-conditioned expert roles as a more effective basis than global routing aggregation for MoE-VLM expert merging.

Evaluating financial AI agents requires criteria aligned with real professional work. Existing rubric methods typically derive criteria from task prompts or model outputs, overlooking tacit standards visible only in practitioner deliverables. We introduce FinProBench, a benchmark for professional financial tasks, and Role-Grounded Rubric Construction (RGRC), a reusable pipeline that derives rubrics from deliverables produced by practitioners in the same role. RGRC comprises four stages: Deliverable Collection, Competency Extraction, Rubric Synthesis, and Validation. Its rubrics capture tacit standards, distinguish quality levels, and transfer across tasks within a role. Before analysis, we classified 57 occupations by deliverable genre into 30 prior-rich conventional roles and 27 prior-sparse role-specialized roles. Across all roles, Prompt-only nearly matches RGRC for conventional roles (89.2% vs. 90.7%), but RGRC substantially outperforms it for role-specialized roles (99.1% vs. 78.0%). This split indicates that prompt engineering can approximate rubrics when conventions are well represented in model priors, while professional grounding is essential for standards beyond those priors. FinProBench is built from 1,723 curated deliverables spanning 57 occupations, 8 financial sub-industries, and 161 deliverable types, and releases an initial evaluation set of 20 complete tasks covering 20 roles in 7 sub-industries. With heterogeneous LLM judges and role-level rubrics, human deliverables rank first on average (73.7 vs. 70.3, 70.2, and 69.6 out of 100), while all four systems show overlapping 95% confidence intervals and complementary strengths. Reusing rubrics at the role level reduces estimated per-task construction effort by 6.7 times relative to authoring each rubric from scratch.

As AI systems are deployed across increasingly diverse social contexts, alignment can no longer be framed as the optimization of a single, unified set of values. Instead, systems must be able to recognize, represent, and respond to multiple legitimate perspectives. This has led to growing interest in pluralistic alignment, which seeks to move beyond one-size-fits-all models of appropriate behaviour. However, current approaches often lack a clear account of how values are socially organized, contested, and coordinated in practice. In this paper, we argue that social theory provides essential conceptual and design resources for addressing these challenges. Drawing on established traditions in sociology, we show how perspectives can be understood as structured by roles, shaped through interaction, and distributed across fields of power and expertise. We translate these insights into concrete implications for AI system design, including role-based representations, structured coordination among perspectives, and context-sensitive evaluation. For agentic systems, this requires aligning not only final outputs, but also the role activations, deliberative traces, aggregation rules, and feedback loops through which those outputs are produced. Our contribution is to reposition pluralistic alignment as a problem of socially grounded coordination rather than output diversification. We outline a design space for systems that engage multiple perspectives in structured and accountable ways, and we identify directions for future work to implement and empirically evaluate these approaches in real-world settings.

Foundation models place language throughout embodied agents, but its presence does not show what it contributes or how well that contribution is grounded. This survey separates these two questions. We define five non-exclusive functional roles for language: Specification, Embodied Representation, Action Orchestration, Grounding Regulation, and Execution Coupling. For each role, we trace the path from linguistic content to its embodied consumer and identify the observations or interventions that can test the claimed responsibility. Applying this framework to the reviewed literature reveals a recurring gap between functional use and evidential support. Interpretable or revised linguistic intermediates may be incorrect, go unused, or fail to affect later behavior. Even when actions are directly conditioned on language, system-level success does not by itself isolate language's contribution. We therefore evaluate grounding claim by claim, asking whether the reported evidence supports the specific responsibility assigned to language. Using role claims rather than architectures as the unit of comparison allows us to compare modular and end-to-end embodied agents without extending conclusions beyond the reported evidence.

Current speech translation systems, including SpeechLLMs, are trained on cleaned text and tend to strip disfluencies like filled pauses and false starts rather than translate them. We show this comes at a cost: disfluencies carry meaning that gets lost when speech is cleaned up. To study this systematically, we introduce Uh-Mazing, a benchmark of human-translated, disfluency-annotated Switchboard speech covering English into eight target languages. Across these languages and several architectures, we find that false starts and self-repairs, not filled pauses or discourse markers, drive most of the translation-quality loss, and that models which fail to preserve a disfluency tend to omit it rather than mistranslate it. We show inference-time decoding can mitigate this without retraining, and release the benchmark and code.

Constructing an executable chemical-process model remains manually intensive. Chemical engineers translate underspecified requests into coupled decisions about unit operations, thermodynamics, streams, specifications, degrees of freedom (DoF), initialization, solver repair, and optimization; one error can invalidate the model. CRAFTS mirrors the staged workflow of chemical engineers by decomposing simulation building into bounded subtasks assigned to seven bounded roles, with deterministic IDAES/Pyomo gates between stages. Given a natural-language request, process flowsheet diagram (PFD) evidence, and curated chemical-engineering knowledge, Input Understanding and Intent recover requirements, constraints, and process semantics; visual, topology, and specification specialists translate them into typed simulator contracts; and Debug and Optimization support bounded repair and eligible optimization. Fine-tuning is applied to the three schema-critical visual, topology, and specification roles, while the remaining roles use untuned Qwen. The resulting VisualGraphIR, TopologyIR, SpecIR, BuildPlan, and SolveReport expose unit, port, thermodynamic, numerical, and execution decisions. Compatible constructors, property packages, and runners are attached only after semantic artifacts pass engineering gates. We introduce OpenIDAES-450, a 450-case IDAES process- simulation dataset, and evaluate the complete seven-role LangChain/LangGraph workflow through solve and eligible optimization on its frozen 82-case held-out split. CRAFTS completes the prescribed validation and execution contract for for 91.5% of cases and achieves unit, stream, and directed-connection F1 scores of 0.815, 0.791, and 0.782. These results demonstrate the effectiveness of role specialization, typed intermediate representations, and deterministic engineering gates for reliable automated process-model construction.

Algorithmic recourse aims to provide individuals with actionable changes to improve their predicted outcomes in high-stakes classification settings, such as loan and mortgage applications. However, most existing approaches focus only on flipping a model's prediction, without accounting for whether the recommended changes lead to genuine improvement in an individual's true qualifications or merely enable strategic gaming of the classifier. Consequently, deployed recourse policies can induce behavioral responses that degrade predictive accuracy and become ineffective after model retraining. In this work, we formalize this failure mode through a causal performative framework for recourse. We model how recourse actions propagate through a structural causal model, capturing interactions among features as well as their effect on the true label. These causal responses induce a non-convex optimization problem, even under standard convex losses. We characterize conditions under which performatively stable solutions exist and can be efficiently computed via simple iterative dynamics. Our analysis reveals that recourse policies that ignore causal structure can induce large, misaligned behavioral responses, whereas causal recourse leads to stable equilibria that reduce incentives for gaming. Experiments on both semi-synthetic and real credit datasets demonstrate that our approach consistently outperforms standard empirical risk minimization while reducing the need for repeated model retraining to accommodate distribution shifts caused by strategic agent behavior.

Relevance is a query-dependent estimate of whether a document or excerpt contains useful evidence. Existing retrieval agents use relevance to select top-$k$ content, but document relevance alone cannot localize, compose, or verify the evidence required by complex questions. Direct Corpus Interaction (DCI) enables such fine-grained operations through grep-style exploration, but its relevance-agnostic search can expose useful clues late and delay convergence. Recent advances use relevance to narrow the corpus into a working space for interaction. Once interaction begins, however, relevance still does not directly guide which documents grep searches first or distinguish informative excerpts from a broad set of matches to let LLMs see them first. We introduce the Relevance-Aware RipGrep Search Agent (RARG), which turns relevance into an execution prior for corpus interaction. RARG provides coarse-to-fine relevance guidance: it orders documents for sequential 'ripgrep' traversal to expose globally relevant clues earlier, initializes promising entry points with query-relevant paragraphs, and reranks grep matches to surface informative excerpts that document-level ranking may otherwise obscure. Across challenging browse question answering and reasoning-intensive retrieval, RARG improves the accuracy--efficiency frontier over retrieval-based and direct-interaction agents. These results demonstrate that relevance-aware interaction enables faster and more reliable search convergence.

Large language models (LLMs) have rapidly and significantly entered scientific workflows, but it remains unclear how their diffusion is associated with changes in scientists' strategies in research directions and team building. We link PubMed Central full text with OpenAlex publication and collaboration histories for 775,323 scientists and analyze CRediT contribution statements from 137,120 multi-author papers. After 2022, scientists increasingly published across more intellectually distant fields and entered fields in which they had not previously worked. These increases in interdisciplinarity and exploration were especially pronounced among established scientists and scientists from non-English-speaking low- and middle-income countries. Authors with stronger AI-writing signals were already more interdisciplinary and exploratory before the widespread adoption of LLMs, and the gap widened further after 2022 compared with authors with weaker AI-writing signals. Scientists' collaboration networks also became more interdisciplinary after 2022. Yet, among authors with stronger AI-writing signals, research interdisciplinarity was less closely tied to the disciplinary diversity of their collaborators. The division of labor within research teams also became more differentiated. Contributors on papers published after 2022 reported narrower role sets on average, coauthors shared fewer roles in common, and their role profiles became less rigid and more fluid. Software and validation roles increased, while conceptual and management roles decreased. These patterns suggest that team members are taking on more distinct responsibilities and may rely less on one another to perform research tasks. Overall, this study indicates that the LLM era coincides with a broader reorganization of scientific exploration, collaboration, and the division of labor.

On-policy distillation (OPD) has become a key paradigm in LLM post-training, yet its training dynamics remain poorly understood. We present a systematic study examining the role, pathologies, and regulations of OPD. We first clarify the role of OPD as an exploration catalyst: it steers the student toward correct reasoning paths via dense token-level guidance, without expanding capability ceiling. We confirm this by showing that prompt diversity matters more than per-problem sampling numbers, and critically, that the effectiveness of OPD hinges entirely on the quality of its guiding signal. This dependency exposes two pathologies that derail exploration. The Student-Teacher Mismatch occurs when a large teacher-student distributional gap causes the guiding signal to misalign with task correctness, steering exploration in counterproductive directions. Length Exploitation arises when the aggregated token-level objective creates length-dependent shortcuts, allowing the student to game the reward landscape through response truncation or redundant padding, exploring degenerate length modes rather than reasoning strategies. To tame these pathologies, we investigate lightweight signal regulations: advantage clipping and log-scale compression, ensuring exploration is guided by faithful signals. Experiments across seven benchmarks demonstrate that these regulations alleviate length exploitation and enable effective distillation, stably surpassing OPD variants and RLVR baselines, thereby confirming that well-regulated signal quality, rather than mere teacher scale, governs successful exploration in OPD.

Attention-based KV cache eviction (H2O and its descendants) compresses the memory-constrained state of a long-context model by ranking tokens on accumulated attention mass, treated here as signal energy, and keeping the heaviest. On schema-dense input streams such as nested JSON, this score acts as a non-stationary filter that disproportionately retains noise: a non-content sink role (delimiters or whitespace) carries an order of magnitude more energy than any content role, and structural KEY tokens are over-retained at roughly 1.8x the rate of the answer-carrying VALUE tokens, collapsing exact-match accuracy from 88% to 0% at a 5% budget as the signal-to-noise ratio of the retained state degrades. A counterfactual experiment establishes that suppressing KEY tokens is the best deployable filter. Our retraining-free, role-conditional allocation over SnapKV's windowed score, governed by a single tuned hyperparameter, closes 63-98% of the H2O gap at sub-20% budgets and, at higher budgets, modestly matches or exceeds full-cache accuracy -- a small, seed-sensitive denoising effect (borderline significant at B=0.50; not distinguishable from zero at B=0.30 over four seeds). A 15 MB linear role probe supplies these labels at negligible inference cost, though matching parser-level downstream accuracy remains open.

Routing policies for multi-model systems are evaluated almost exclusively on task accuracy and inference cost. We argue that two properties, orthogonal to performance, determine whether routing is meaningful. First, the society of actors must be behaviourally differentiated: if all actors respond identically, routing is vacuous. Second, the routing policy must be stable: surface-form variants of a query should be assigned to the same actor. High task accuracy is compatible with violating both properties, since a router can operate over a redundant society or assign queries inconsistently, preventing specialisation regardless of performance. We adapt Hierarchic Social Entropy (HSE) to language-model societies and introduce a perturbation-based robustness metric to diagnose these failure modes. Applied to EmbedLLM and RouterBench, we find that HSE exhibits strong diminishing returns, suggesting that a curated subset of fewer than ten agents recovers most available diversity in a large pool -- a practical coreset heuristic for society design. We further find that KNN routers gain accuracy from specialist societies but collapse in robustness under perturbation, while prompted routing remains stable across all perturbation types -- illustrating that accuracy and meaningfulness can sharply diverge.

Which properties of a partially trained network are causally portable to a different, independently trained network? Single-trajectory interventions show necessity within one run, not portability across runs. We introduce cross-trajectory chimera interventions: given two runs from different seeds, we split each weight vector into a norm and a unit direction, recombine one run's norm with the other's direction, and continue training. On two modular-arithmetic tasks that grok, the components dissociate. Direction carries a transferable, donor-specific circuit identity: implanting a donor's direction at the recipient's norm drives the run to the donor's circuit in 40/40 cases, while an angle-matched random control yields no shift. The transfer is threshold-like, and its location is predicted by the recipient's norm, separating perfectly by norm class over all 20 pairs (joint permutation probability 1.9e-4). Norm carries only a modest, distributed delay effect and no identity signal. An adaptive bisection procedure localizes the threshold to +/-1/64. Direction indexes which solution a trajectory approaches; norm governs how susceptible that identity is to being overwritten.

Large language models (LLMs) can generate BPMN process models from natural-language descriptions, yet supervised fine-tuning (SFT) limits their output quality to the patterns present in the training data. Reinforcement learning (RL) can optimize beyond this ceiling using external quality measures, but how the reward function should be designed when quality is multi-dimensional remains unexplored. We present a systematic investigation of reward function design for RL-based process model generation, training two LLM families (Llama3.1 8B, Qwen2.5 14B) under 48 configurations using Group Sequence Policy Optimization with rewards derived from an automated evaluation framework comprising 38 metrics across syntactic, pragmatic, and semantic quality. Three findings emerge. First, RL significantly improves pragmatic and syntactic quality while preserving semantic fidelity, reducing output variability by more than sixfold. Second, equal reward weighting consistently outperforms targeted weighting: emphasizing a specific dimension fails to improve it and can collapse the model into a low-quality mode. Third, design choices interact with model architecture in non-trivial ways: the invalidity penalty is essential for one model but irrelevant for the other, and SFT initialization is indispensable for one architecture but counterproductive for another. These results demonstrate that reward composition is a primary determinant of optimization outcomes, with effects as large as the decision to apply RL itself. The findings generalize to any structured generation task where quality is assessed along multiple automated dimensions. We release our implementation and experimental code at https://github.com/chlauer99/RL_for_process_modeling.

Chamber music, as a highly precise multi-part interactive system, contains a logic of "role assignment and dynamic interaction" that provides an extremely valuable blueprint for exploring human-computer collaborative composition paradigms. Addressing the lack of role perception capabilities in existing deep music generation models during polyphonic interactions, this paper conducts an interdisciplinary analysis of Haydn's String Quartet in D Major, The Lark (Op. 64, No. 5). We propose a novel research path: "Classical Morphology Qualitative Analysis-Electroacoustic Quantitative Measurement-Machine Representation Reconstruction." The study first utilizes auditory analysis to dissect the counterpoint morphology of the leading voice and the underlying groove in the first movement. Subsequently, it introduces spectrum and dynamic feature analysis tools from a Digital Audio Workstation (DAW) to translate subjective auditory perception into objective, measurable physical parameters. Building on this, the paper introduces a fundamentally new approach to low-level computer feature extraction: completely abandoning the traditional mechanical quantization grid, introducing Event-based Timestamps to record the duration of micro-timing, and transforming acoustic features into an independent "Role-Aware Encoding" as an aesthetic heuristic mechanism (a phenomenological anchor). This study not only completes the logical loop spanning classical analysis, electronic music mapping, and AI symbolic generation but also establishes a profound theoretical foundation-from the perspectives of interactive aesthetics and media philosophy-for constructing human-computer collaborative music systems imbued with "social attributes" and "otherness awareness."

We present our systems for SemEval-2026 Task 10 (PsyCoMark), addressing conspiracy marker extraction (Subtask 1) and document-level conspiracy detection (Subtask 2). For marker extraction, we formulate the task as multi-label span classification over enumerated candidate spans, using IoU >= 0.95 positive labeling, hard-negative sampling, and containment-based non-maximum suppression (NMS) with boundary-aware span representations. Document classification is modeled independently using a sequence classifier with label smoothing and a stratified train-validation split. Analysis shows that entity-like roles (Actor, Victim) are detected robustly, while abstract roles (Action, Effect, Evidence) remain sensitive to boundary criteria. On the official test set, our systems rank 7th in Subtask 1 (0.2251 macro F1) and 11th in Subtask 2 (0.7694 weighted F1).

Why do intelligent systems need to perform explicit symbolic reasoning? Computer science has traditionally regarded symbolic reasoning as a defining component of intelligence. Yet the remarkable success of modern foundation models raises a fundamental question: if increasingly capable AI systems can operate with little explicit symbolic reasoning, what role do symbolic methods actually play? This article argues that explicit symbolic reasoning is not a fundamental property of intelligence, but a computational consequence of operating on simplified models of reality. We propose the Compression Principle: every computational model is a simplified representation of reality, and explicit symbolic reasoning compensates for information omitted during model construction. From this principle, we derive the Modeling--Reasoning Trade-off: as computational models preserve richer representations of the world, the need for explicit symbolic reasoning correspondingly decreases. This perspective provides a unified explanation for both the historical success of symbolic methods and the remarkable effectiveness of modern foundation models. Paradoxically, the same development makes symbolic methods increasingly important for humans. As intelligent systems become more capable and more opaque, symbolic representations increasingly serve as interfaces through which humans specify requirements, verify behavior, regulate autonomous systems, and establish trust. We therefore argue that the future of symbolic methods lies not primarily as the computational engine of intelligent systems, but as the symbolic interface between increasingly capable AI systems and the humans who build, govern, and depend upon them.

Decentralised federated learning, based on peer-to-peer communication, is increasingly proposed for on-device training of machine learning models, promising a privacy-preserving, communication-efficient training process with no risk of single-point failure. However, the role of structural and temporal inhomogeneities in such fully decentralised settings remains poorly understood. Here, we investigate their effects when model parameters are locally averaged during aggregation. We show that the decentralised federated learning process is governed, both in the early phase and the late, stationary limit, by the same dynamics as a lazy random-walk diffusion process on temporal networks. Based on this mapping, we demonstrate that the typical experimental scenario used in decentralised federated learning leads to unrealistically rapid convergence because of ignoring the temporal and structural inhomogeneities inherent in the communication network. We analyse real-world temporal networks and find that inhomogeneities most often dramatically slow down diffusion, hence the convergence process.

Several SLOG test categories explicitly involve directional distinctions (modifier position shifts, argument extraction positions), yet AM-Parser, the previous SOTA, uses an AM algebra whose operations do not encode direction. We redesign the symbolic backend around CCG directed types (deterministic CKY + single linear decoder, 30K learnable parameters). Under the same BERT-base encoder, the system achieves 75.9$\pm$6.4% LF exact match, surpassing AM-Parser (70.8$\pm$4.3%). Per SLOG's own category groupings, gains are highly directional: the CCG system outperforms AM-Parser on all 5 position-shift categories (+29.9pp), while AM-Parser outperforms on all 6 recursive-depth categories. Replacing the encoder with DeBERTa-v3-large yields 90.7$\pm$4.9%, with the largest encoder gains in recursive-depth categories, complementary to directionality's gains. Directional representations shift the bottleneck from the symbolic layer (AM-Parser's 0% category ceiling) to the neural layer, which improves with encoder upgrades.

Autonomous customer-service agents are shifting from conversational interfaces toward operational execution roles: they retrieve firm records, apply service policies, and execute backend writes such as refunds, cancellations, exchanges, order modifications, and reservation changes. This shift creates a service-control problem: firms must keep routine service fast and low-friction while preventing operational errors on requests where customer instructions, policy constraints, firm records, and backend writes interact. We propose a difficulty-routed service-control architecture that asks when service agents should reconsider before acting. A lightweight router keeps routine sessions on a low-cost baseline path and routes operationally coupled sessions to an escalated workflow. The escalated path uses conflict-aware communication and write-triggered reconsideration to concentrate deliberation and safeguards before consequential backend writes, rather than applying additional control uniformly across all service sessions. We evaluate the architecture on human-verified retail and airline tasks from $τ^{2}$-bench. In retail, the method improves reliability consistently on service requests with operational conflict. Routing evidence shows that stronger control is directed toward conflicted requests rather than broadly applied to routine ones. Dialogue and tool-use profiles suggest that gains do not come from indiscriminate interaction expansion or broader tool chains; instead, added turns and tool calls support evidence gathering, write separation, and pre-write reconsideration. Case-level evidence shows that the escalated workflow preserves fallback plans, binds retrieved records to the correct action, sequences writes, and decomposes multi-entity requests. Airline results extend the same service-control logic to reservation operations.

Mixture-of-Experts (MoE) architectures have recently been extended with role-based mechanisms for interpretability. This is typically done by assigning semantic roles to individual expert components, for example roles like synergy, redundancy, and uniqueness in multimodal settings. However, whether such structural role decomposition preserves explanation faithfulness of the overall architecture remains largely underexplored. We hypothesize that inter-expert representation overlap weakens effective role separation and degrades attribution-based faithfulness, even when semantic roles are explicitly defined. To address this limitation, we introduce representation-level decorrelation regularization to explicitly reduce inter-expert similarity in latent space. Using representation decorrelation objectives, we encourage clearer specialization among experts by minimizing representation overlap. Our experiments show that across multiple multimodal benchmarks, this separation consistently improves explanation faithfulness, as measured by comprehensiveness, sufficiency, and their Area Over the Perturbation Curve (AOPC) summaries, while preserving task performance. We further show that these improvements are not limited to role-based architectures such as Interpretable Multimodal Interaction-aware MoE (I2MoE). Similar trends are observed in a standard sparse MoE baseline, suggesting that representation-level separation may provide a more general mechanism for enhancing explanation faithfulness in MoE systems. Overall, our findings suggest that structural role decomposition alone may be insufficient to guarantee faithful explanations and that representation-level separation helps improve explanation faithfulness. To support reproducibility, the source code and supplementary material are publicly available at https://github.com/dut0817/FL-I2MoE_Decor.

Role-semantic assignments provide priors over how heterogeneous agents may coordinate, but cooperative MARL systems instead settle on conventions through decentralized, non-stationary learning, with no guarantee that the resulting structure matches those priors. We study this translation gap between theory-informed role expectations and learned coordination structure through a diagnostic combining a role-routing matrix, formation sensitivity ($Δ_{\max}$), and gradient/occlusion attribution across three-role MiniGrid and SMACv2 (Terran) environments. We show that label-conditioned attention produces substantially more concentrated and role-specific routing than flat MLP baselines, remains stable under 3v3--9v9 scaling, transfers zero-shot across team sizes, and is invariant to ally-slot padding. A 5-seed re-evaluation shows partial alignment between learned conventions and designer-specified priors while revealing where small-n noise can manufacture apparent strategic divergence. We present these results as an empirical framework for measuring coordination structure in cooperative MARL rather than as a new equilibrium concept or causal explanation.

Team roles offer an interpretable lens on collaboration, yet computational studies of roles often rely on domain-specific personas or data-driven clustering rather than theory-grounded taxonomies. We operationalize a taxonomy of eight communication roles grounded in education literature and annotate a corpus of 6,307 Slack messages from 55 students across 18 teams in a semester-long computer science course project. We evaluate whether LLMs can approximate expert labels, enabling scalable, taxonomy-driven role annotation. Using these role labels, we characterize role dynamics over teams' lifecycles, finding that different roles peak at different moments and that students enact a more diverse set of roles as projects progress. To evaluate the utility of our role constructs, we use them to predict peer recognition, outperforming lexical, conversational, and LLM-prompting baselines. To assess generalizability beyond the educational context, we apply the same role constructs to a public dataset (DeliData) to predict team performance improvement after deliberation, again exceeding prior performance.

Several theoretical works have tried to explain the adversarial vulnerability of deep neural networks through properties of high-dimensional geometry. However, the assumptions underlying these works are rarely examined empirically, and systematic evidence remains limited. In this work, we present a systematic study of the role of input dimensionality in both the emergence and the targeted control of adversarial examples. We first analyse the scope and limitations of existing theoretical frameworks based on concentration of measure, showing that real image classes exhibit strong empirical localization, beyond what such theories typically assume. We then conduct an extensive empirical evaluation across hierarchical image datasets spanning a wide range of input dimensionalities and diverse neural architectures. Our results consistently show that adversarial examples become easier to construct as dimensionality increases. We also investigate how input dimensionality affects the additional difficulty of crafting targeted adversarial examples. In particular, we provide theoretical arguments showing that high-dimensional geometry implies that enforcing a specific target label entails only a limited additional distortion compared to untargeted attacks. We corroborate this insight through extensive experiments, demonstrating that the gap between targeted and untargeted perturbations remains small and further narrows as input dimensionality increases. While, taken together, our findings establish high input dimensionality as a fundamental factor underlying the emergence and targeted control of adversarial examples, whether this phenomenon primarily arises from the interplay between high-dimensional geometry and data distributions or from the architectural properties of deep neural networks remains an open question.

Artificial Intelligence (AI) is rapidly reshaping the nature of work in software development, transforming user roles, workflows, and collaboration patterns across enterprise platforms. This qualitative study investigates how AI alters professional responsibilities within the context of SAP's Business Technology Platform (BTP), combining expert interviews (n=20) and a participatory workshop (n=24). The results reveal substantial shifts in day-to-day tasks and roles in the development domain, characterized by increasing automation of operational tasks, expanding human-AI collaboration, and growing reliance on agentic AI systems. The study further identifies significant implications for existing user-role frameworks, such as the BTP User Type Matrix, which requires adaptation as the workforce is undergoing significant role specific changes. Collectively, these findings highlight a workforce landscape in transition and underscore the need for revised role taxonomies, new governance and oversight functions, and updated design approaches for AI-native enterprise software systems.

Federated Learning (FL) has emerged as a key paradigm for privacy-preserving collaborative model training across distributed and heterogeneous data sources. By keeping raw data local, FL addresses data confidentiality concerns, yet it does not resolve the opacity of modern machine learning models. In parallel, Explainable Artificial Intelligence (XAI) has gained attention for improving transparency, trust, and accountability, particularly in high-stakes domains. Their intersection has given rise to Federated Explainable Artificial Intelligence (FedXAI) paradigm, which aims to jointly satisfy privacy and explainability requirements. This survey provides a systematic review of FedXAI, highlighting the transition of explainability from a post-hoc tool to an integral component of the FL lifecycle. We show how explainability supports aggregation, personalization, robustness, coordination, and system-level decision making. To organize the literature, we introduce a taxonomy that classifies FedXAI methods by the role of explainability, model and explainer types, explanation scope, integration level, FL settings, and data heterogeneity. We review approaches ranging from model-agnostic explanations to interpretable federated models and explainability-aware aggregation mechanisms. We also examine evaluation practices and discuss the lack of standardized benchmarks and metrics for measuring explanation quality, stability, privacy leakage, and computational overhead. Finally, we identify key challenges, including explainability under non-IID data, explanation-centric security threats, communication-efficient XAI, continual FedXAI, and the integration of domain knowledge and regulatory constraints. By consolidating existing work and identifying key gaps, this survey serves as a reference framework for designing trustworthy, transparent, and privacy-preserving federated AI systems.

Modern language models increasingly adopt hybrid architectures that combine full attention with efficient attention modules, such as sliding-window attention (SWA) and recurrent sequence mixers. However, how these efficient modules shape model capabilities remains poorly understood. To address this gap, we conduct a systematic analysis across hybrid architectures from three perspectives: scaling behavior, mechanism analysis, and architecture design. First, from a scaling perspective, we find that efficient-attention design primarily affects how fast long-context capability emerges, while different hybrids eventually converge to comparable long-context performance under sufficient training. Second, mechanistically, we show that long-range retrieval is mainly carried by full attention, whereas efficient attention shapes its optimization trajectory. This explains a counter-intuitive phenomenon we call Large-Window Laziness: larger SWA windows can delay the formation of retrieval heads in full-attention layers. Third, guided by this mechanism, we show that applying NoPE to only the full-attention layers of a small-window SWA hybrid substantially improves long-context performance with negligible impact on short-context performance.

Collecting participants' lived experiences is central to design research. Focus groups are uniquely valuable because participants not only share individual accounts but also respond to one another, surfacing comparison, disagreement, and collective sensemaking. However, focus groups are resource-intensive and highly sensitive to facilitation: moderators must probe for specificity, balance participation, manage topic flow, and sustain psychological safety, and subtle facilitation choices can shape what becomes salient. Recent HCI work and commercial meeting tools show that generative AI can scaffold live conversation through prompting, turn regulation, thematic mapping, and real-time summarization. Yet UXR teams lack a clear map of what these capabilities mean in focus groups and what methodological risks they introduce. We synthesize AI supports for live conversation and translate them into a focus-group-specific playbook organized by AI role (tool, co-host, host) and modality (text, voice, embodied).We synthesize prior work on AI-supported live conversation and propose a focus-group-specific playbook of AI supports organized by role (tool, co-host, host) and modality (text, voice, embodied). We characterize interactional trade-offs and identify open questions for evaluating AI-supported focus groups as methodological configurations.

Conditioning a language model on additional context, such as feedback on a previous attempt, typically improves its response. Self-distillation trains the model to retain this improvement when the context is not present. The method works by matching the model's output distribution under two settings: a student that sees only the question, and a self-teacher that also sees the context. What the model learns therefore depends on what context the self-teacher receives, yet the design of this context remains largely unexplored. We study context design for self-distillation by training a solver on feedback from a frozen critic. We compare three conditions: (i) a binary reward (GRPO), (ii) the reference solution, and (iii) a step-by-step critique aligned to the solver's reasoning trace. Step-aligned critique yields the largest gains, outperforming GRPO by 16.11 points and reference-solution-conditioned self-distillation by 5.27 points (Avg@12). Per-token advantage analysis reveals why: step-aligned feedback targets only the tokens where reasoning fails, leaving correct behavior intact. Conditioning on the reference solution, by contrast, pressures the model to change its behavior at every token (even correct steps) because an alternative derivation inevitably differs in phrasing and approach. This suggests that structural alignment between feedback and the solver's reasoning is a key driver of self-distillation effectiveness.

Selecting which companies to approach is a central challenge in business-to-business (B2B) sales, where decisions are often based on manual research and fragmented information sources. Modern B2B sales platforms centralize company records and use learned company embeddings to support tasks such as recommending and prioritizing potential clients. In this study, we investigate whether enriching these company embeddings with Semantic knowledge from DBpedia improves downstream interaction-prediction performance, within a pipeline that integrates structured company attributes and text embeddings deployed on a real B2B platform. We evaluate the learned embeddings on a downstream interaction prediction task using real user feedback data from the platform. Results show that DBpedia enrichment improves downstream performance, with gains observed on ranking and discrimination metrics.

Post-training compression is essential for deploying large language models (LLMs) under tight resource constraints. Tensor decompositions have emerged as a promising direction, offering compact parameterizations well suited to Transformer weight structures. However, existing studies evaluate these methods in narrow settings, leaving unclear whether tensorization is effective at large-scale deployment. We systematically evaluate tensor compression across dense and MoE architectures, establishing performance trade-offs grounded in both empirical analysis and theoretical analysis. We identify a fundamental mismatch between the shared subspaces assumed by tensor decompositions and the heterogeneous representations learned by modern LLMs, thereby delineating their practical limits and clarifying their viable role in large-scale deployment. The code is available at https://github.com/brain-lab-research/TT-LLM.

In the contemporary epoch of multilingual education, learning idioms provides a fascinating gateway towards creativity, cultural values, historical context, and diverse perspectives inherent to various linguistic traditions. This paper showcases the navigation of retaining figurative and cultural semantics in low-resource Southeast Asian languages such as Hindi, Bengali, and Thai, where culturally rich idioms pose significant obstacles for computational modeling and cross-linguistic transfer due to their deep metaphorical complexity. To tackle such complexity, we present Varnika, a reconstructed multimodal idiom corpus comprising 3,533 multilingual idioms, enriched with seven idiomatic tones aligned with both textual and visual representations. Additionally, to infer informative idiomatic understanding, we introduce a Hybrid Mixture-of-Experts (HybridMoE) framework that embeds multiple idiomatic expert opinions while mitigating expert sparsity by integrating outputs from both selected and unselected experts through controlled hybridization, further augmented with Idiomatic Property Signals via masked multimodal embeddings. To analyze the performance across multiple dimensions, we propose the IDIO-TONE and Idiomatic Validation Score, a three-stage evaluation pipeline measuring (i) literal translation fidelity, (ii) visual-semantic alignment, and (iii) idiomatic meaning retention. Empirical evaluations highlight that HybridMoE achieves 5--6% performance gains across advanced vision language models, demonstrating improved representation of figurative language and culturally embedded meaning in multilingual multimodal settings

Role-playing agents(RPAs) are widely used to steer large language models(LLMs) toward role-consistent behavior, yet existing benchmarks mainly evaluate surface-level fidelity and offer limited insight into decision making under role-alignment value conflicts. To address this gap, we introduce RoleCDE, the first benchmark designed to evaluate RPAs under structured conflicts between role-specific values and alignment-oriented constraints. RoleCDE formulates role-aware decision making as cognitive dilemma scenarios, jointly evaluating role-scenario grounding, value conflict resolution, and decision tendencies. The benchmark is constructed at scale, covering approximately 8k diverse role profiles and scenarios and nearly 24k dilemma instances across three difficulty levels and eight role categories. Evaluation of several mainstream LLMs reveals a "Role Value Decoupling" phenomenon, where agents systematically default to alignment-and morality-consistent decisions rather than role-specific values when the two conflict, even under explicit role conditioning. This behavior is largely invariant to dilemma difficulty but varies substantially across role categories. We further show that RoleCDE-based fine-tuning effectively mitigates this decoupling by improving value trade-off reasoning, while preserving general role-playing fidelity and general reasoning performance. Code is available at: https://github.com/rabbitrose/RoleCDE.

Positional encoding (PE) is widely viewed as necessary for transformers to process ordered sequences: without them, the next-token map appears permutation-invariant in its context tokens. This intuition underlies all prior universality results, which rely on positional information to prove that transformers with chain-of-thought can perform arbitrary computation, i.e., they are Turing complete. We revisit this belief in the regime most relevant to long-form reasoning, where generation proceeds through a finite sliding context window. Our opening perception is that the window mechanism itself (mildly) breaks the permutation symmetry. To distill and precisely capture the degree of this added expressiveness, we introduce an abstract autoregressive model, the HIST model, in which each update depends only on constant-size internal state and the token-count histogram within the current window. We prove that this HIST model is Turing complete by showing that the evolution of the window can reveal the token that has just left the window, which suffices to simulate Turing-complete Post machines. We then construct a sliding-window transformer over a constant-size token alphabet, without PE, and show that it can simulate the HIST model. Our result demonstrates that positional encodings are not indispensable for transformers to perform universal computation: The window sliding itself already breaks permutation symmetry and captures sufficient positional information.

It has been proposed that the brain integrates flexible, computationally expensive cortical processing with simpler, lower-cost subcortical mechanisms to achieve resource-efficient performance greater than that of either system alone. Despite the allure of this perspective, satisfying theoretical frameworks that explore this hypothesis are still limited. We extend existing frameworks in which a model-based module and model-free module learn in tandem by explicitly constraining the memory resources of the model-based module, and investigate the impact of this constraint in a simple decision-making setting. Memory constraints naturally give rise to strategies for allocating memory resources. We evaluate the performance of different strategies in different situations and demonstrate that when the rewarded states change often, it can be advantageous for the model-based module to focus its memory resources not on exploiting the current reward, but on capturing general structure of the environment. This work provides a theoretical foundation for a functional dissociation between cortical and subcortical systems during learning: the cortex supports general structure learning, while subcortical circuits specialize in reward-based learning. We further detail how these hypotheses can be tested on experimental data.

Reverse Kullback-Leibler (RKL) divergence is widely favored over forward KL (FKL) in large language models (LLM) distillation, yet this preference is largely based on comparisons that omit the temperature $τ$, overlooking its central role in softening teacher distributions and improving knowledge transfer. In this work, we revisit temperature in LLM distillation and show that it fundamentally changes the comparison between FKL and RKL. Our analysis reveals an asymmetric effect: temperature substantially enriches FKL with non-dominant token signals, whereas it mainly rescales RKL gradients, causing FKL to benefit much more from $τ$ scaling than RKL. This asymmetry overturns the standard empirical conclusion: although RKL outperforms FKL at $τ=1$, FKL consistently surpasses RKL at higher temperatures across instruction-following benchmarks. Moreover, the impact of temperature is not limited to FKL; it improves a broader family of distillation objectives, enabling simple KL-based methods to achieve competitive performance against recent state-of-the-art LLM distillation approaches.

LLM agents are increasingly deployed as multi-role teams, where tasks are divided across specialized roles such as planner, executor, and verifier. In these systems, cost and accuracy are no longer properties of a single model: they depend on which model fills each role and where it is hosted, including API, self-hosted, and hybrid deployment. Existing agentic benchmarks typically evaluate fixed models or fixed agent configurations, and therefore offer limited guidance for cost-accuracy-optimal deployment. We introduce AgentCARD, a role-aware benchmark suite for evaluating LLM agent teams across role assignment and deployment mode. AgentCARD combines a role-decomposed evaluation harness, a unified API/self-hosted cost model, Pareto-frontier analysis, and a Shapley-based diagnostic for identifying role bottlenecks. Our evaluation shows that heterogeneous teams consistently occupy the cost-accuracy frontier. They improve accuracy by up to $44\%$ over cost-equivalent homogeneous teams, or match the strongest homogeneous team at up to $12\times$ lower per-task cost through hybrid deployment. We further find that the best role assignment is domain-dependent: some domains are planner-bottlenecked, while others are executor-bottlenecked. Finally, AgentCARD extends beyond planner--executor teams to workflows with additional roles such as verification, and supports continual evaluation as new domains and team structures emerge. Our code is released at: https://github.com/Auto-CAP/AgentCAP

Language models are increasingly being deployed for conversational support in informal caregiving contexts, where interactions often extend beyond information-seeking: caregivers seek emotional reassurance, guidance, and help, while navigating uncertain, relationally complex care decisions. Yet most safety evaluations assess model behavior under generic prompts, leaving a critical question unexamined: does a model's safety profile change with its support role? We study this by operationalizing four expert-reviewed support roles grounded in social support theory: Inform, Coach, Relate, and Listen, and comparing them against two baseline controls: a basic prompting condition and a retrieval-augmented generation (RAG) condition. We evaluate across three language models (GPT-4o-mini, Llama-3.1-8B-Instruct, and MedGemma-1.5-4b-it) on 5,000 real-world queries from online Alzheimer's Disease and Related Dementias (ADRD) communities. We find that the LLM's support role systematically shapes both the prevalence and composition of interactional risks. Furthermore, a human evaluation study reveals a perceived quality--safety tension: more directive, information-oriented roles are rated as more helpful and trustworthy despite exhibiting elevated interactional risk profiles. We release ~90,000 support role-conditioned model responses with risk annotations as an ecologically grounded resource for research on safer LLM-mediated conversational support.

Safety detection models require examples of HHH (Helpful, Harmless, Honest)-violating outputs for robust generalization, however such examples are scarce. Activation Steering (AS) has emerged as a data-efficient method for generating target-concept-aligned responses. We investigate whether AS can generate high-quality training datasets for downstream classifiers, a question that remains untested. We present a two-fold study with intrinsic and extrinsic evaluation across $4$ concepts $\times\,2$ models $\times\,4$ steering methods. Intrinsically, beyond the field-standard rubric of steering success (concept alignment) and coherence, we introduce sample- and set-level diversity as a quality axis previously absent from the literature, and find that increasing steering strength reduces response diversity. Extrinsically, we replace HHH-violating examples in the available training data with steered generations and fine-tune detection classifiers. AS-generated data results in a better classifier than the prompting-generated data on $3$ of $4$ concepts. However, only $41$ of $136$ AS configurations outperform prompting, indicating that downstream utility lies in a narrow regime that jointly satisfies success, coherence, and diversity. The harmonic mean of these three axes correlates with downstream AUROC more consistently across concepts than success and coherence alone, providing a practical heuristic target for practitioners tuning AS hyperparameters. Together, our results highlight the potential of AS in synthetic data generation for improving safety detection and identify diversity as a critical, previously overlooked axis for tuning AS.

Role-based LLM multi-agent systems need adaptive role pools, yet adapting such systems is not merely a matter of prompt optimization: roles often carry structural obligations, including capability coverage, message compatibility, validation, final-answer aggregation, and parser-compatible output protocols. Existing systems either fix the role inventory and lose adaptivity, or allow unconstrained generation to induce role drift, removing structurally necessary roles and breaking answer contracts. We formulate this as contract-preserving role evolution, requiring every committed edit to preserve five structural contracts (capability, communication, validation, aggregation, output protocol). We instantiate this formulation in SERO, a Self-Evolving Role Orchestration framework that evolves a typed role-card pool through credit-guided retrieval, a credit-ranked communication DAG with a protected terminal aggregator and conditional validator repair, and a contextual-bandit controller whose LLM-proposed edits are committed only when they preserve the contracts and improve task score. Experiments on real-world reasoning benchmarks across three LLM backbones confirm the value of contract-preserving role evolution.

Reinforcement Learning with Verifiable Reward (RLVR) is empirically shown to notably enhance the reasoning performance of large language models (LLMs), particularly in mathematics and programming. However, the mechanistic role of Sample Difficulty in RLVR remains poorly understood. In this paper, we investigate RLVR through the lens of difficulty-wise and one-sample analysis. We find that sample difficulty has a non-monotonic effect on RLVR: easy and medium-difficulty problems yield the strongest and most stable reasoning improvements, whereas overly hard problems often provide weak learning signals, induce degenerate behaviors such as answer repetition or skipping necessary computation, and can ultimately degrade the model's pre-existing capabilities. Beyond the obverse of response, we further analyze the model's internal feature dynamics using Temporal Sparse Autoencoders (T-SAE). Easy problems mainly reinforce direct-answer and basic-computation features while suppressing deliberative-reasoning features; hard problems activate reasoning-related features but become useful only when successful trajectories are sampled; medium-difficulty problems provide a more balanced signal, strengthening both computation and multi-step reasoning features. Motivated by these findings, we propose difficulty-adaptive strategies for hard-sample utilization, using backward-reasoning reformulation and T-SAE-guided training signals to improve reward density and credit assignment during RLVR. Overall, our results identify sample difficulty as a key factor governing both the optimization dynamics and representation evolution of RLVR.

Prompt-based personality control is a key technique for designing large language model (LLM) dialogue agents that behave consistently across social contexts. However, specifying Big Five personality traits (BFTs) in a prompt does not ensure that the intended traits are expressed in generated utterances. This paper investigates this mismatch from an interactionist perspective, viewing personality expression as a context-dependent outcome shaped by the interplay between trait specification and situational factors. We analyze how perceived BFT expression in LLM-generated dialogue is influenced by three prompt factors: personality traits, dialogue roles, and expressive styles. Using a factorial design that combines six personality conditions, three roles, and three expressive-style conditions, we generate 1,080 LLM-agent dialogues in each of English and Japanese. We then evaluate the target agent's utterances using an LLM-as-a-judge framework to estimate expressed Big Five traits. The results show that expressed personality is shaped not only by explicit trait specification, but also by dialogue role and expressive style. These effects are trait-specific: dialogue role strongly influences Openness, expressive style substantially shapes Conscientiousness and Agreeableness, and explicit trait specification dominates Neuroticism. Even without explicit personality-trait specification, social and expressive conditions induce distinct personality-like impressions. Cross-linguistic comparisons show broadly similar patterns between English and Japanese dialogues, with noticeable differences only under specific combinations of personality, role, and expressive style. These findings suggest that personality control in LLM agents should be understood not as a direct consequence of trait prompting, but as a context-dependent process involving personality specification, social role, and expressive style.

Online recruitment platforms require recommendation methods capable of retrieving relevant job opportunities from large and heterogeneous collections of job postings. Keyword-based search is efficient and interpretable, but it may fail to retrieve relevant postings when equivalent roles are expressed using different terminology. This study presents a metadata-driven job recommendation system that combines TF-IDF lexical matching, Sentence-BERT semantic retrieval, query-aware filtering, optional Cross-Encoder re-ranking, and explanation generation. The proposed system utilizes structured metadata fields including job title, company name, location, seniority level, job function, employment type, and industry without relying on full job descriptions or user interaction histories. Experiments conducted on a cleaned LinkedIn job posting dataset containing 31262 records demonstrate that the best hybrid configuration achieved a Precision at 10 score of 0.8032 and an nDCG at 10 score of 0.9496. Under the internal evaluation protocol, Cross-Encoder re-ranking improved Precision at 10 from 0.7896 to 0.7948 and nDCG at 10 from 0.9666 to 0.9739. These findings indicate that lexical and semantic retrieval techniques can be effectively combined to provide explainable job recommendations when only structured metadata is available.

In-context learning (ICL) is often motivated by the intuition that demonstrations help because they provide correct input-output examples. However, we reveal a counterintuitive phenomenon: correctness does not guarantee exemplar utility, and some correct demonstrations can even reduce ICL accuracy. To study this correctness-utility gap, we introduce task-preserving perturbations, where only the exemplar input is changed, while the example remains a correct instance of the same task. Concretely, each perturbed exemplar is assigned the target induced by the task mapping. This framework covers both label-updating perturbations, where task-relevant semantics change and targets are recomputed, and stricter target-preserving perturbations, where the original target remains valid. We formalize the resulting failure mode as contextual evidence shift: task-preserving perturbations can change the effective mixture of evidence used by the model for contextual inference, thereby separating exemplar correctness from exemplar utility. Across sentiment classification, logical reasoning, and math word problems, we find that task-preserving perturbed demonstrations can substantially degrade ICL performance, especially for smaller models, harder tasks, and higher perturbation ratios. Our results show that robust ICL requires evaluating not only whether demonstrations are correct, but also how they influence contextual inference. Code is available at https://github.com/Chenghao-Qiu/Task-Preserving-ICL.

Large language models (LLMs) have achieved strong performance in code generation, but most methods rely on autoregressive decoding without global planning, often leading to locally coherent yet globally suboptimal solutions (e.g., failing test cases or inefficient complexity). While recent approaches such as Chain-of-Thought (CoT) and multi-agent systems (MAS) introduce planning, their limited role specialization and coordination hinder performance on complex tasks. To address the challenges of coordination and specialization in multi-agent code generation, we propose Collaborative Reinforcement Code (CoRe-Code), a framework for role specialized LLM agents that enhances inter-agent coordination to generate more accurate and efficient code. CoRe-Code adopts a simple Planner-Coder paradigm, where the Planner produces high-level plans and the Coder executes them to generate code. We further introduce a collaboration-aware reinforcement learning stage based on Group Relative Policy Optimization (GRPO) to enhance role specialization and alignment. Experiments show that CoRe-Code outperforms a wide range of existing RL-based and multi-agent methods. In addition, we demonstrate that CoRe-Code can generalize to other multi-agent frameworks (e.g., Retrieval and Debugging agents), highlighting its flexibility and scalability. We evaluate CoRe-Code on multiple benchmarks of varying difficulty using three base models. Compared to existing baselines, the results show consistent improvements in accuracy, while also achieving higher efficiency in terms of execution time and memory usage, demonstrating the effectiveness and practicality of CoRe-Code.

Role prompts of the form As X, do Y admit a clean linear decomposition at one specific site in the residual stream: the prompt-to-answer transition -- the last prompt token together with the first two generated tokens -- in an early/mid layer band. There, persona and task contribute through partially orthogonal additive directions. Forming a pure persona effect $Δ_X$, a pure task effect $Δ_Y$, and substituting $h_{BB} + Δ_X + Δ_Y$ for the clean residual yields downstream output within a small KL of clean on Gemma-2-2B-IT and Qwen-2.5-{1.5B, 3B}-Instruct, across a 12-cell short grid and a 48-cell long-persona grid, with persona-specific behavioral markers preserved. The natural inference from this additive structure is that the role prompt can be compressed into a single cached residual vector. \emph{We show it cannot.} Injecting the cached additive prediction -- or even the oracle clean residual $h_{XY}$ -- into a baseline host prompt with the persona text removed does not approach the clean long-persona target, at one site or at many layers. Persona-conditioned multi-token generation flows through attention back to the persona-text positions throughout the prompt, which no residual at one site reproduces. Local additivity in the residual stream does not imply prompt compressibility. The additive structure at the prompt-to-answer transition supports interpretability and fine-grained steering of persona or task contributions; persona-conditioned behavior across the full continuation depends on a distributed prompt/KV mechanism that local activation arithmetic does not displace.

An increasing number of news and research articles report that AI adoption is allowing professionals to blur and extend the boundaries of their corporate roles. With the goal of understanding how work processes might be changing in an AI-forward company, we interviewed 24 product-focused individuals at a large technology firm about how AI has impacted their own work, their work within their product team, and their professional interactions. Our conversations suggest that AI is not only changing formal role responsibilities and collaborations between those roles, but also changing informal cultural practices like professional mentoring that are key to helping professionals settle in their positions, stay engaged with their work, and grow their careers. Some of these changes are positive, such as smoother collaboration between peers, but other changes are more nuanced and put the typical career growth opportunities, like receiving feedback from professional networks and promoting leadership and mentorship, at risk. We propose steps that AI companies can take to make the invisible work more visible. Additionally, we propose efforts that individuals and leaders can take to support their colleagues through AI transformation while preserving healthy company cultures that support diverse thinking, collaboration, and informal interactions.

Cloud-hosted LLM driver agents provide useful semantic judgments, but their inference latency exceeds stepwise vehicle-control windows. Learned world models predict futures, but they usually keep future generation and action selection inside large coupled loops. We present SteinsGateDrive, a latency-decoupled planner-runtime architecture in which the worldline metaphor from the eponymous story names one plausible consequence of an intervention: the LLM selects counterfactual driving futures before the final control instant, and a runtime reuses the selected forecast only while safety contracts remain valid. The generator builds three world-line roles: alpha nominal ego-conditioned futures, beta interaction counterfactuals around nearby vehicles, and gamma hazard-stress futures such as braking, cut-ins, or blocked corridors. The selected branch becomes a typed StrategicForecast with horizon, validity/abort conditions, fallback, and authority. On a within-subject, matched-seed normal-highway protocol with 10 seeds and 20 steps, GPT-5.4 mini reduces effective lag from +3.07 s at 1-second horizon to -0.01 s at 4-second horizon while preserving the measured no-collision safety boundary. The architecture's safety contribution comes from the atom-predicate runtime check, not from the drift score, which functions as a refresh-frequency knob.

Artificial intelligence (AI) has achieved extraordinary capabilities despite lacking many of the conceptual and scientific foundations associated with mature disciplines. Unlike traditional sciences, where reliable technology typically emerges from theoretical understanding, modern AI has progressed largely through performance-driven iteration and "alchemical" experimentation. This tension motivates a systematic analysis of AI through the lens of rigor. We introduce a three-part framework consisting of conceptual rigor (clarifying foundational concepts), epistemic rigor (establishing scientific understanding), and operational rigor (ensuring reliable performance and deployment). Using this framework, we analyze competing conceptions of intelligence and understanding, the strengths and limitations of the empirical approach to deep learning, the power and pitfalls of benchmarks, and the obstacles to theory development posed by modern AI systems. We argue that the distinctive trajectory of AI arises from how forms of rigor interact across paradigms, resulting in the primacy of operational rigor in modern deep learning. This perspective helps explain both AI's rapid advances and its persistent uncertainties, while clarifying the challenges involved in transforming AI into a mature science and reliable technology.

Social roles shape expectations, priorities, and judgments, yet it remains unclear how large language models (LLMs) associate occupational identities with broader cultural value patterns. Prior work used nationality-based cultural prompting to study how LLM responses to value-survey questions align with human cultural benchmarks. In this paper, we extend that framework by replacing cultural prompting with occupational prompting to examine how professional-role cues influence value-survey responses in open-weight LLMs. Using a survey-grounded evaluation pipeline based on questions from the Integrated Values Surveys, we project model responses into the two-dimensional Inglehart--Welzel cultural space. We prompt open-weight LLMs to answer questions under occupational identities such as accountant, teacher, engineer, and nurse, and then analyze how these occupation-conditioned responses are positioned on the cultural map. Our results show that when open-weight LLMs are prompted with occupations rather than national identities, their responses remain within a broadly Western-leaning region of the cultural map. However, different occupations introduce shifts within this region, producing distinct occupational skews. This indicates that occupational prompts are not treated as neutral role labels, but instead elicit structured value patterns. These findings extend survey-based evaluation of cultural bias beyond nationality-based prompting and provide a framework for studying how occupational personas shape value expression in LLMs.