Search research

Integrating expressive generative policies, such as flow-matching models, into offline reinforcement learning (RL) allows agents to capture complex, multi-modal behaviors. While Q-learning with Adjoint Matching (QAM) stabilizes policy optimization via the continuous adjoint method, it remains inherently bound to the fixed behavior distribution. This dependence induces a \textit{popularity bias} that can suppress high-reward actions in low-density regions, and creates a \textit{support binding} that restricts off-manifold exploration. Existing workarounds, such as appending \textit{residual} Gaussian policies, often re-introduce the expressivity bottlenecks associated with unimodal distributions. In this work, we propose \textit{Maximum Entropy Adjoint Matching} (ME-AM), a unified framework that addresses these limitations within the continuous flow formulation. ME-AM incorporates two mechanisms: (1) a Mirror Descent entropy maximization objective that mitigates the popularity bias to facilitate the extraction of optimal policies from offline datasets, and (2) a \textit{Mixture Behavior Prior} that broadens the geometric support to encompass out-of-distribution high-reward regions. By exploring this extended geometry, ME-AM identifies robust actions while preserving the absolute continuity of the generative vector field. Empirically, ME-AM demonstrates competitive or superior performance compared to prior state-of-the-art (SOTA) methods across a diverse suite of sparse-reward continuous control environments.

Unsupervised pretraining has driven empirical advances in goal-conditioned reinforcement learning (GCRL), but its theoretical foundations remain poorly understood. In particular, an influential class of methods, mutual information skill learning (MISL), discovers behaviorally diverse skills that can later be used for downstream goal-reaching. However, it remains a theoretical mystery why skills learned through MISL should support goal-reaching. A subtle challenge is that both GCRL and MISL are umbrella terms: different GCRL tasks use distinct criteria for measuring goal-reaching performance, while different MISL methods optimize distinct notions of behavioral diversity. We address this challenge and unify GCRL and MISL as instances of control maximization. We identify three canonical GCRL formulations and prove that they are fundamentally inequivalent: they can induce incompatible optimal policies even in the same environment. Nevertheless, they all share a common interpretation: a well-performing goal-conditioned policy is one whose future trajectory is highly sensitive to the commanded goal, with the precise notion of sensitivity determined by the GCRL formulation. Noting that MISL objectives can be understood as measures of skill-sensitivity akin to goal-sensitivity, we show that MISL objectives are bounded by formulation-specific downstream goal-sensitivities. These bounds establish a precise correspondence between MISL methods and downstream GCRL tasks: for every GCRL formulation, there exists a matching MISL objective for which more diverse skills afford greater downstream goal sensitivity. Our results thus lay a theoretical foundation for RL pretraining and have important practical implications, such as suggesting which pretraining objectives to use when a user cares about a specific class of downstream tasks.

In multimedia application scenarios, images captured under low-illumination conditions often lead to lower accuracy in visual perception tasks compared to those taken in well-lit environments. To tackle this challenge, we propose AMIEOD, an image enhancement-enabled object detection framework for low-illumination scenes, where the two tasks are jointly optimized in a detection performance-oriented manner. Specifically, to fully exploit the information in poorly lit images, a Multi-Experts Image Enhancement Module (MEIEM) is proposed, which leverages diverse enhancement strategies. On this basis, aiming to better align the MEIEM with the detection task, we propose a Detection-Guided Regression Loss (DGRL) that utilizes the detection result to decide the regression target. Moreover, to dynamically select the most suitable enhancement strategy from MEIEM during inference, we construct an Expert Selection Module (ESM) guided by the proposed Detection-Guided Cross-Entropy (DGCE) loss, which formulates the optimization of ESM as a classification task. The improved method is well-matched with current detection algorithms to improve their performance in dim scenes. Extensive experiments on multiple datasets demonstrate that the proposed method significantly improves object detection accuracy in low-illumination conditions. Our code has been released at https://github.com/scujayfantasy/AMIEOD

This paper studies runtime monitoring for persistent surveillance by autonomous robots when the autonomy stack is a black box. The environment is partitioned into finitely many parts, each carrying an uncertainty state that decreases when observed and increases otherwise. We model the closed loop as a state-dependent hybrid system with linear parameter varying dynamics and design a monitor based on an invariant computed offline. As this invariant is typically hard to obtain for large to-be-surveyed spaces, we propose a compositional monitor obtained by decentralized computation of low-dimensional invariant sets for each uncertainty region, and checking their conjunction online. Under common independence assumptions, the compositional monitor is sound and complete with respect to the full-system invariant. The approach is applied in a case study with a real robot persistently monitoring a labyrinth, emphasizing its applicability in practice.

As spoken dialogue systems expand beyond traditional assistant roles to encompass diverse personas -- such as authoritative instructors, uncooperative merchants, or distracted workers -- they require distinct, human-like turn-taking behaviors to maintain psychological immersion. However, current full-duplex systems often default to a rigid, overly accommodating ``always-yield'' policy during overlapping speech, which severely undermines character consistency for non-submissive roles. Evaluating alternative, persona-specific turn-taking strategies through empirical user studies is challenging because building real-time full-duplex test environments requires substantial engineering overhead. To address this, we present PersonaKit (PK), an open-source, low-latency web platform for the rapid prototyping and evaluation of conversational agents. Using intuitive JSON configurations, researchers can define personas, specify probabilistic interruption-handling behaviors (e.g., yield, hold, bridge, or override), and automatically deploy comparative A/B surveys. Through an in-the-wild evaluation with 8 distinct personas, we demonstrate that PersonaKit provides an extensible, end-to-end framework for studying complex sociolinguistic behaviors in next-generation spoken agents.

We present 3D Surface Splatting (3DSS), the first differentiable surface splatting renderer for physically-based inverse rendering from multi-view images. Our central insight is that the surface separation problem at the heart of surface splatting admits a direct formulation in terms of the reconstruction kernels themselves. From this foundation we derive a coverage-based compositing model whose per-layer opacity arises directly from the accumulated Elliptical Weighted Average reconstruction weight, yielding anti-aliased silhouettes and informative visibility gradients at sparsely covered edges. Combined with forward microfacet shading under co-optimized HDR environment lighting and density-aware adaptive refinement, 3DSS jointly recovers shape, spatially-varying BRDF materials, and illumination. Because the optimized representation is a set of oriented surface samples, it bridges natively to mesh-based workflows via surface reconstruction from oriented point cloud methods. We evaluate 3DSS against mesh-based, implicit, and Gaussian-splatting baselines across geometry reconstruction, novel-view synthesis, and novel-illumination relighting.

Future networking systems are envisioned to become part of an agentic AI-native ecosystem in which a vast number of heterogeneous and specialized AI agents cooperate seamlessly to fulfill complex user requirements in real time. However, traditional networking paradigms are characterized by a rigid decoupling of communication and computation, which often leads to significant inefficiencies in large-scale agentic AI networking (AgentNet) systems. Emergent communication offers a novel solution by enabling autonomous agents that support task-specific signaling protocols for information exchange and collaborative coordination. In this paper, we consider a multi-agent emergent communication framework, tailored for semantic-aware AgentNet systems in which the user's semantic intent can be automatically detected, inferred, and linked to a set of sub-tasks to be assigned to a set of agents. We investigate how communication and signaling protocols can emerge among collaborative agents with computationally bounded intelligence under stringent bandwidth constraints. Our proposed framework, called SANEmerg, is designed to facilitate the emergence of communication for collaborative task fulfillment while adhering to the physical limits of AgentNet. SANEmerg incorporates a bandwidth-adaptable importance-filter that dynamically prioritizes the transmission of higher-contribution message dimensions, ensuring robust performance in bandwidth-limited environments. Furthermore, SANEmerg integrates a complexity-regularizer grounded in the Minimum Description Length (MDL) principle to facilitate the emergence of computationally bounded signaling. Evaluated via an AgentNet prototype and extensive experimentation, SANEmerg demonstrates significant performance improvements over state-of-the-art solutions, achieving superior task accuracy while significantly reducing bandwidth and computational overhead.

Group-relative RL training (GRPO) samples a small group of parallel rollouts for every training prompt and uses their within-group reward spread to compute per-trajectory advantages. In agentic environments each rollout is a long multi-turn dialogue with one LLM call per step, so this multi-sample multiplier dominates the total training cost. When every rollout of a prompt ends with the same reward, the group has zero reward variance and contributes no gradient, so the extra rollouts add no information; such groups are common in practice (typically around 40% of all groups), so the wasted-compute fraction is substantial rather than marginal. Existing methods filter such groups at the prompt level, either after their rollouts are paid for or before any rollout begins, but both decide without using information that becomes available during the rollout itself. We instead ask whether the in-group divergence between the partial trajectories at an intermediate step can already predict that the group will be zero-variance: when the parallel rollouts have already converged on the same action prefix, the group is on track to produce a single reward, and we can stop early. We propose a one-parameter gate that stops a group when the mean pairwise prefix edit distance between its partial action sequences falls below a threshold. On a 60-iteration on-policy GRPO run on ALFWorld with Qwen2.5-7B, averaged over four random seeds, the gated arm finishes 10.7% faster in wall-clock (bootstrap 95% CI excludes 0) and shifts held-out success rate on 50 unseen tasks by +2.5 pp, with the held-out gain tracing to a measurable reduction in zero-advantage gradient-batch dilution. Code is available at https://github.com/zhiyuanZhai20/selective-rollout.

Explainable Artificial Intelligence (XAI) is increasingly rec ognized as essential for deploying machine learning systems in safety critical environments. In Automatic Target Recognition (ATR), where models operate on image, video, radar, and multisensor data, high pre dictive performance alone is insufficient. Model decisions must also be interpretable, reliable, and suitable for validation. This paper presents a structured evaluation of explainability methods in the context of safety-critical ATR systems: We identify major XAI paradigms, including saliency-based, attention-based, and surrogate ap proaches, as well as recent detection-aware extensions. Based on this, we formalize explainability as an assurance-oriented assessment problem, introduce a taxonomy, and assess these methods with respect to four key dimensions: interpretability, robustness, vulnerability to manipula tion, and suitability for validation and verification. The analysis identifies systematic limitations of current post-hoc explanation methods. In par ticular, we derive critical failure modes such as spurious explanations, instability under perturbations, and overtrust induced by visually con vincing outputs. These findings indicate that widely used XAI techniques may be insufficient for safety-critical deployment. Finally, we discuss implications for ATR systems and outline directions toward more robust, causally grounded, and physically informed explain ability methods. Our results emphasize the need to move beyond visually plausible explanations toward approaches that support reliable decision making and system-level assurance.

Long-range electrostatics and polarization remain central obstacles to extending machine learning interatomic potentials (MLIPs) to ionic, polar, and interfacial systems. Here we introduce a semi-local framework for learning electrostatics from energies and forces using polarizable atomic multipoles. Local equivariant descriptors predict environment-dependent latent monopoles, dipoles, and quadrupoles, while residual non-local charge transfer and polarization are captured by non-self-consistent linear response in induced charges and dipoles. Across four diverse benchmarks and four short-range MLIP architectures, the multipole hierarchy and response terms systematically improve potential energy surface accuracy, with the largest gains in systems where long-range effects are essential. More importantly, physically meaningful electrical responses emerge without direct supervision. The learned latent multipoles yield accurate Born effective charge tensors and infrared spectra in close agreement with experiments. The induced-dipole extension introduces new capabilities: it predicts polarizabilities and thereby enables semi-quantitative Raman spectra for bulk water and hybrid MAPbI$_3$ perovskite, as well as the essential features of the surface-specific vibrational sum-frequency generation spectrum at the water-air interface. In ferroelectric HfO$_2$, the predicted electrical response also captures LO-TO splitting and polarization switching. This systematically improvable, physically transparent framework enables MLIPs trained on standard energy and force labels to predict polarization-sensitive observables.

Knee osteoarthritis (KOA) is among the musculoskeletal disorders that considerably restrict joint mobility, cause severe chronic pain and impact negatively on quality life. It is one of the persistent health issues worldwide. Generally, subjectivity and inter-observer variability undermine conventional practices and evaluation process that are adopted to address such health issues. Hence precise and timely diagnosis would be one of the effective ways for the assessment of its severity. This paper proposes an automated diagnostic approach for severity grading of KOA by blending a deep learning convolutional neural network (CNN) with a device-based inference platform powered by TensorFlow Lite. It proposes a model based on the ResNet-18 convolutional neural network. The designed model is trained on publicly available database. Through a transfer learning approach obtained knee images are first classified into five Kellgren-Lawrence (KL) grades. Further the developed model is optimised. During the training of the model test accuracy of 94.48% with stable convergence has been achieved. Subsequently the optimised model transformed into a lightweight TensorFlow Lite format, facilitating seamless deployment on resource-constrained devices. The designed model is capable enough to operate in the environment having no continuous internet connectivity. Also, an auxiliary Large Language Model (Gemini-2.0-flash) is applied to generate structured interpretive findings like potential symptoms, risk factors, and preventive majors etc. The LLM component functions as interface without influencing the classification process. The proposed model articulates the feasibility of an on-device, interpretable decision-support tools for early diagnosis and improve accessibility to Artificial Intelligence (AI)-assisted knee screening tool.

We study auto research as a closed empirical loop driven by external measurement. Each submitted trial carries a hypothesis, an executable code edit, an evaluator-owned outcome, and feedback that shapes the next proposal. The output is not a generated paper or a single model checkpoint, but an auditable trajectory of proposals, code diffs, experiments, scores, and failure labels. We instantiate this loop with specialist agents that partition recipe surfaces and share measured lineage across trials. The central empirical finding is that lineage feedback lets agents turn evaluator outcomes, including crashes, budget overruns, size failures, and accuracy-gate misses, into later program-level recipe edits rather than one-shot suggestions. Across 1,197 headline-run trials plus 600 Parameter Golf control trials after one-time setup and launch, humans did not choose proposals, edit recipes, override scores, or repair failed trials during the search. In the three headline runs, the same submitted-trial loop reduces Parameter Golf validation bpb by $0.81\%$, raises NanoChat-D12 CORE by $38.7\%$, and reduces CIFAR-10 Airbench96 wallclock by $4.59\%$, with each task measured by its own external evaluator and legality checks. The trace includes a strict architecture-domain audit of 157 headline-run submissions and program rewrites such as a NanoChat attention-kernel path change. Within this scope the loop autonomously writes code, submits experiments, absorbs feedback, applies and combines known techniques inside each environment, and improves public starting recipes.

Cooperative inference across independently deployed machine learning models is increasingly desirable in distributed environments, as there is a growing need to leverage multiple models while keeping their data and model parameters private. However, existing cooperative frameworks typically rely on sharing input data, model parameters, or a common encoder, which limits their applicability in privacy-sensitive or cross-organizational settings. To address this challenge, we propose Consensus Embedding-based Federated Inference (CE-FI), a framework that enables pretrained models to cooperate at inference time without sharing model parameters or raw inputs and without assuming a common encoder. CE-FI introduces two components: a Consensus Embedding (CE) layer that maps heterogeneous intermediate representations into a common embedding space, and a Cooperative Output (CO) layer that produces predictions from these embeddings. Both layers are trained using shared unlabeled data only, so the cooperative stage does not require additional labeled data. Experiments on image classification benchmarks -- CIFAR-10 and CIFAR-100 -- under diverse non-IID conditions show that CE-FI consistently outperforms solo inference and performs comparably to conventional methods that require stronger sharing assumptions. Additional evaluations on text and time-series tasks indicate applicability beyond image classification, although performance depends on the ensemble strategy. Further analysis identifies representation alignment as the primary bottleneck.

Recent advances in large language models (LLMs) have significantly improved language-driven 3D content generation, but most existing approaches still treat scene generation and user interaction as separate processes, limiting the adaptability and immersive potential of interactive multimedia systems. This paper presents a unified framework that closes the loop between language-driven 3D scene generation and immersive user interaction. Given natural language instructions, the system first constructs structured scene representations using LLMs, and then optimizes spatial layouts via reinforcement learning under geometric and semantic constraints. The generated environments are deployed in a virtual reality setting to facilitate HRI-in-the-loop, where user interactions provide continuous feedback to align generated content with human perception and usability. By tightly coupling generation and interaction, the proposed framework enables more responsive, adaptive, and realistic multimedia experiences. Experiments on the ALFRED benchmark demonstrate state-of-the-art performance in task-based scene generation. Furthermore, qualitative results and user studies show consistent improvements in immersion, interaction quality, and task efficiency, highlighting the importance of closed-loop integration of generation and interaction for next-generation multimedia systems. Our project page can be found at https://proj-showcase.github.io/h3ds/.

Recent advances in foundation models have transformed LLMs from passive conversational systems into autonomous agents capable of reasoning and tool execution. While these capabilities unlock substantial practical value, they also introduce new security risks, as adversaries can manipulate agents into performing harmful actions in real-world environments. Existing defense strategies mitigate such threats but frequently struggle to balance safety and utility, resulting in over-refusal of benign user requests. To mitigate this trade-off, we propose SafeHarbor, a novel framework designed to establish precise decision boundaries for LLM agents. Unlike static guidelines, SafeHarbor extracts context-aware defense rules through enhanced adversarial generation. We design a local hierarchical memory system for dynamic rule injection, offering a training-free, efficient, and plug-and-play solution. Furthermore, we introduce an information entropy-based self-evolution mechanism that continuously optimizes the memory structure through dynamic node splitting and merging. Extensive experiments demonstrate that SafeHarbor achieves state-of-the-art performance on both ambiguous benign tasks and explicit malicious attacks, notably attaining a peak benign utility of 63.6% on GPT-4o while maintaining a robust refusal rate exceeding 93% against harmful requests. The source code is publicly available at https://github.com/ljj-cyber/SafeHarbor.

Large language model (LLM)-based coding assistants have made substantial progress, yet most systems remain reactive, requiring developers to explicitly formulate their needs. Proactive coding assistants aim to infer latent developer intent from integrated development environment (IDE) interactions and repository context, thereby reducing interaction overhead and supporting more seamless assistance. However, research in this direction is limited by the scarcity of large-scale real-world developer behavior data. Existing studies therefore often rely on LLM-simulated IDE traces, whose fidelity to real development behavior remains unclear. In this paper, we investigate this simulation-to-reality gap through a large-scale empirical study. We collect real IDE interaction traces from 1{,}246 experienced industry developers over three consecutive days using a custom Visual Studio Code extension, and construct paired LLM-simulated traces for controlled comparison. Our analysis shows that simulated traces differ substantially from real traces in behavioral diversity, temporal structure, and exploratory patterns. Based on the collected data, we introduce \textbf{ProCodeBench}, a real-world benchmark for proactive intent prediction. Experiments with representative LLMs, retrieval-augmented methods, and agentic baselines show that current approaches remain far from reliable under real IDE traces, suggesting that simulation-based evaluation can overestimate real-world performance. Finally, our training study shows that simulated data cannot replace real data, but can complement it when used before real-world fine-tuning. These findings highlight the importance of real developer behavior data for evaluating and training proactive coding assistants.

Large Language Model (LLM)-based agents have fundamentally reshaped artificial intelligence by integrating external tools and planning capabilities. While memory mechanisms have emerged as the architectural cornerstone of these systems, current research remains fragmented, oscillating between operating system engineering and cognitive science. This theoretical divide prevents a unified view of technological synthesis and a coherent evolutionary perspective. To bridge this gap, this survey proposes a novel evolutionary framework for LLM agent memory mechanisms, formalizing the development process into three stages: Storage (trajectory preservation), Reflection (trajectory refinement), and Experience (trajectory abstraction). We first formally define these three stages before analyzing the three core drivers of this evolution: the necessity for long-range consistency, the challenges in dynamic environments, and the ultimate goal of continual learning. Furthermore, we specifically explore two transformative mechanisms in the frontier Experience stage: proactive exploration and cross-trajectory abstraction. By synthesizing these disparate views, this work offers robust design principles and a clear roadmap for the development of next-generation LLM agents.

The proliferation of large language models (LLMs) in educational settings has paradoxically undermined the cognitive processes they purport to support. Students increasingly outsource critical thinking to AI assistants that generate polished text on demand, resulting in measurable cognitive debt and diminished argumentative reasoning skills. We present Prober.ai, a web-based writing environment that inverts the conventional AI-tutoring paradigm: rather than generating or rewriting student text, the system constrains an LLM (Gemini 3 Flash Preview) through persona-specific system prompts and structured JSON output schemas to produce only targeted, inquiry-based questions about argumentative weaknesses. A two-phase interaction architecture -- Challenge and Unlock -- implements a pedagogical friction mechanism whereby revision suggestions are gated behind mandatory student reflection. The system's design is grounded in Toulmin's argumentation theory, research on peer feedforward questioning mechanisms, and evidence on AI-supported feedback in writing instruction. A functional prototype was developed in 36 hours during the NY EdTech Hackathon (March 2026), where it was awarded second place. We describe the system architecture, the prompt engineering methodology for constraining LLM output to pedagogically aligned JSON schemas, and discuss implications for scalable, cognition-preserving AI integration in writing education.

LLM agents that operate over long context depend on external memory to accumulate knowledge over time. However, existing methods typically store each observation as a single deterministic conclusion (e.g., inferring "API~X failed" from temporary errors), even though such observations are inherently partial and potentially ambiguous. By committing to one conclusion and discarding uncertainty, these methods introduce self-reinforcing error: the agent acts on the stored conclusion, never revisits alternatives, and reinforces the conclusion over time. To address this issue, we propose BeliefMem, which shifts the memory paradigm from committing to a single conclusion per observation to retaining multiple candidate conclusions with their probabilities. Concretely, BeliefMem stores the candidate conclusions as separate memory entries, each carrying a probability that is updated via Noisy-OR rules as new observations arrive. At retrieval, all candidates surface together with their probabilities, keeping alternatives visible to the agent. Since each conclusion in memory retains its probability, BeliefMem preserves the uncertainty that the deterministic paradigm discards, enabling the agent to act with high confidence on well-evidenced knowledge while retaining the capacity to update its confidence when new evidence arrives. Empirical evaluations on LoCoMo and ALFWorld benchmarks show that, even with limited data, BeliefMem achieves the best average performance, remarkably outperforming well-known baselines. More broadly, such probabilistic memory produces substantial gains and explores a new direction for agent memory in partially observable environments.

Federated learning (FL) operates in heterogeneous environments, where variations in data distributions and asymmetric model design often result in negative transfer. While federated knowledge distillation (FKD) avoids direct model parameter sharing, existing methods typically rely on public datasets or assume that transferred knowledge is uniformly reliable, which limits their robustness in practice. This paper presents FedeKD, a reliability-aware FKD framework that makes sample-wise trust estimation an explicit component of knowledge transfer, without relying on additional public data. Each client maintains a high-capacity private model for local learning and a lightweight shared proxy model for cross-client knowledge exchange. During training, proxy models are aggregated on the server to form a global proxy, which is then used to guide updates of the private models. At the core of FedeKD is an energy-based gating mechanism that converts task-specific private-proxy disagreement into sample-wise trust weights for backward distillation. This mechanism enables sample-wise weighting of knowledge transfer, where the proxy model contributes more to reliable samples while down-weighting unreliable ones. Extensive experiments on six real-world datasets demonstrate that FedeKD significantly reduces negative transfer under heterogeneous settings while maintaining strong predictive performance.

We present AgenticRAG, a practical agentic harness for retrieval and analysis over enterprise knowledge bases. Standard RAG pipelines place significant burden of grounding on the search stack, constraining the language model to a fixed candidate set chosen deep in the retrieval process. Our approach reduces this overdependence by layering a lightweight harness on top of existing enterprise search infrastructure, equipping a reasoning LLM with search, find, open, and summarize tools enabling the model to iteratively retrieve information, navigate within documents, and analyze evidence autonomously. On three open benchmarks we observe substantial gains: $49.6\%$ recall@1 on BRIGHT (+21.8 pp over the best embedding baseline), 0.96 factuality on WixQA ($+13\%$ relative improvement), and $92\%$ answer correctness on FinanceBench--within 2 pp of oracle access to true evidence. Ablation studies show that the most significant factor is the shift from single-shot retrieval to agentic tool use ($5.9\times$ improvement), while multi-query search and in-document navigation contribute to both quality and efficiency. We present various design choices in our agentic harness that were informed by pre-production deployments. Our results demonstrate its suitability for real-world enterprise production environments.

Scaling LLM-based embodied agents from text-only environments to complex multimodal settings remains a major challenge. Recent work identifies a perception-reasoning-decision gap in standalone Vision-Language Models (VLMs), which often overlook task-critical information. In this paper, we introduce PRISM, a framework that tightly couples perception (VLM) and decision (LLM) through a dynamic question-answer (DQA) pipeline. Instead of passively accepting the VLM's description, the LLM critiques it, probes the VLM with goal-oriented questions, and synthesizes a compact image description. This closed-loop interaction yields a sharp, task-driven understanding of the scene. We evaluate PRISM on the ALFWorld and Room-to-Room (R2R) benchmarks. We show that: (1) PRISM significantly outperforms state-of-the-art image-based models, (2) our Interactive goal-oriented perception pipeline yields systematic and substantial gains, and (3) PRISM is fully automatic, eliminating the need for handcrafted questions or answers.

Enterprise agents increasingly operate inside scoped retrieval systems, delegated workflows, and policy-constrained evidence environments. In these settings, access control can be enforced correctly while the system still produces an answer that appears complete even though material evidence lies outside the caller's authorization boundary. This paper introduces Partial Evidence Bench, a deterministic benchmark for measuring that failure mode. The benchmark ships three scenario families -- due diligence, compliance audit, and security incident response -- with 72 tasks total, ACL-partitioned corpora, oracle complete answers, oracle authorized-view answers, oracle completeness judgments, and structured gap-report oracles. It evaluates systems along four surfaces: answer correctness, completeness awareness, gap-report quality, and unsafe completeness behavior. Checked-in baselines show that silent filtering is catastrophically unsafe across all shipped families, while explicit fail-and-report behavior eliminates unsafe completeness without collapsing the task into trivial abstention. Preliminary real-model runs show model-dependent and scenario-sensitive differences in whether systems overclaim completeness, conservatively underclaim, or report incompleteness in an enterprise-usable form. The benchmark's broader contribution is to make a governance-critical agent failure measurable without human judges or contamination-prone static corpora.

We present ZAYA1-8B, a reasoning-focused mixture-of-experts (MoE) model with 700M active and 8B total parameters, built on Zyphra's MoE++ architecture. ZAYA1-8B's core pretraining, midtraining, and supervised fine-tuning (SFT) were performed on a full-stack AMD compute, networking, and software platform. With under 1B active parameters, ZAYA1-8B matches or exceeds DeepSeek-R1-0528 on several challenging mathematics and coding benchmarks, and remains competitive with substantially larger open-weight reasoning models. ZAYA1-8B was trained from scratch for reasoning, with reasoning data included from pretraining onward using an answer-preserving trimming scheme. Post-training uses a four-stage RL cascade: reasoning warmup on math and puzzles; a 400-task RLVE-Gym curriculum; math and code RL with test-time compute traces and synthetic code environments built from competitive-programming references; and behavioral RL for chat and instruction following. We also introduce Markovian RSA, a test-time compute method that recursively aggregates parallel reasoning traces while carrying forward only bounded-length reasoning tails between rounds. In TTC evaluation, Markovian RSA raises ZAYA1-8B to 91.9% on AIME'25 and 89.6% on HMMT'25 while carrying forward only a 4K-token tail, narrowing the gap to much larger reasoning models including Gemini-2.5 Pro, DeepSeek-V3.2, and GPT-5-High.

Offline reference trajectories for active target tracking are needed both for building multi-modal tracking datasets and for benchmarking online tracking planners under repeatable conditions. We present Track A star (TA star), an offline search-based trajectory planner that targets the visibility-aware target tracking objective on a discretized four-dimensional spatio-temporal grid (x, y, z, t). TA star combines a layered Directed Acyclic Graph (DAG) search with three engineering optimizations: cross-time obstacle distance caching against a Bounding Volume Hierarchy (BVH), per-layer beam pruning, and a configurable multi-ray visibility evaluator. TA star employs a beam-pruned heuristic search on this discrete graph to efficiently find high-quality tracking trajectories. While it trades strict theoretical optimality for practical scalability, our empirical results demonstrate robust, near-baseline visibility performance at a fraction of the computational cost. On a 1000-scenario stress test across eight CARLA Optimized maps, TA star converges on all scenarios and completes in 45 s using 32 workers; on a 248-scenario controlled comparison against an unoptimized priority-queue A star baseline (BinaryHeap implementation) under identical scenario inputs and a 5 x 10^6 expansion cap, TA star reduces mean planning time by 23.0x and worst-case planning time by 11.8x, while raising convergence from 56.9% to 100%. On the n=141 baseline-converged subset, TA star changes average visibility by only -0.15 percentage points (pp), with no scenario exceeding a 5 pp drop. We position TA star as a practical offline reference planner under these specific conditions, with limitations and failure cases discussed for environments such as Town07 dense vegetation.

Retrieval-Augmented Generation (RAG) and agentic AI systems are increasingly prevalent in enterprise AI deployments. However, real enterprise environments introduce challenges largely absent from academic treatments and consumer-facing APIs: multiple tenants with heterogeneous data, strict access-control requirements, regulatory compliance, and cost pressures that demand shared infrastructure. A fundamental problem underlies existing RAG architectures in these settings: retrieval systems rank documents by relevance--whether through semantic similarity, keyword matching, or hybrid approaches--not by authorization, so a query from one tenant can surface another tenant's confidential data simply because it scores highest. We formalize this gap and analyze additional shortcomings--including tool-mediated disclosure, context accumulation across turns, and client-side orchestration bypass--that arise when agentic systems conflate relevance with authorization. To address these challenges, we introduce a layered isolation architecture combining policy-aware ingestion, retrieval-time gating, and shared inference, enforced through server-side agentic orchestration. This approach centralizes security-critical operations--tool execution authorization, state isolation, and policy enforcement--on the server, creating natural enforcement points for multitenant isolation while allowing client-side frameworks to retain control over agent composition and latency-sensitive operations. We validate the proposed architecture through an open-source implementation in OGX, a vendor-neutral framework that implements an OpenAI-compatible, open-source Responses API with server-side multi-turn orchestration. We evaluate it empirically and show that ABAC gating eliminates cross-tenant leakage while introducing negligible overhead.

Deep search has become a crucial capability for frontier multimodal agents, enabling models to solve complex questions through active search, evidence verification, and multi-step reasoning. Despite rapid progress, top-tier multimodal search agents remain difficult to reproduce, largely due to the absence of open high-quality training data, transparent trajectory synthesis pipelines, or detailed training recipes. To this end, we introduce OpenSearch-VL, a fully open-source recipe for training frontier multimodal deep search agents with agentic reinforcement learning. First, we curated a dedicated pipeline to construct high-quality training data through Wikipedia path sampling, fuzzy entity rewriting, and source-anchor visual grounding, which jointly reduce shortcuts and one-step retrieval collapse. Based on this pipeline, we curate two training datasets, SearchVL-SFT-36k for SFT and SearchVL-RL-8k for RL. Besides, we design a diverse tool environment that unifies text search, image search, OCR, cropping, sharpening, super-resolution, and perspective correction, enabling agents to combine active perception with external knowledge acquisition. Finally, we propose a multi-turn fatal-aware GRPO training algorithm that handles cascading tool failures by masking post-failure tokens while preserving useful pre-failure reasoning through one-sided advantage clamping. Built on this recipe, OpenSearch-VL delivers substantial performance gains, with over 10-point average improvements across seven benchmarks, and achieves results comparable to proprietary commercial models on several tasks. We will release all data, code, and models to support open research on multimodal deep search agents.

We present a dual-barrier control barrier function (CBF) safety filter for real-time, safety-critical velocity control of holonomic robots operating in incrementally built occupancy grid maps. As a robot explores an unknown environment, unmapped regions introduce irreducible uncertainty, since obstacle geometry beyond the explored frontier is unknown, making entry into such regions a source of collision risk, especially with front-facing sensors. To address this, we enforce two constraints: avoidance of mapped obstacles and restriction from unexplored regions. Both constraints are derived analytically from the occupancy grid's signed distance field, yielding a closed-form safety filter that requires only a small linear system solve per cycle. On resource-constrained platforms such as the Raspberry Pi, where SLAM and planning already consume significant compute, the low overhead of the proposed filter preserves resources. An adaptive gain schedule relaxes the frontier constraint in information-rich regions and tightens it in well-mapped areas, improving exploration efficiency while maintaining safety. The filter operates in velocity space as a minimally invasive correction and composes with arbitrary nominal controllers, including learning-based methods. Hardware flight experiments on a PX4-controlled quadrotor demonstrate zero collisions across multiple indoor runs.

Behavior Cloning (BC) has emerged as a highly effective paradigm for robot learning. However, BC lacks a self-guided mechanism for online improvement after demonstrations have been collected. Existing offline-to-online learning methods often cause policies to replace previously learned good actions due to a distribution mismatch between offline data and online learning. In this work, we propose Q2RL, Q-Estimation and Q-Gating from BC for Reinforcement Learning, an algorithm for efficient offline-to-online learning. Our method consists of two parts: (1) Q-Estimation extracts a Q-function from a BC policy using a few interaction steps with the environment, followed by online RL with (2) Q-Gating, which switches between BC and RL policy actions based on their respective Q-values to collect samples for RL policy training. Across manipulation tasks from D4RL and robomimic benchmarks, Q2RL outperforms SOTA offline-to-online learning baselines on success rate and time to convergence. Q2RL is efficient enough to be applied in an on-robot RL setting, learning robust policies for contact-rich and high precision manipulation tasks such as pipe assembly and kitting, in 1-2 hours of online interaction, achieving success rates of up to 100% and up to 3.75x improvement against the original BC policy. Code and video are available at https://pages.rai-inst.com/q2rl_website/

In offline-to-online reinforcement learning (O2O-RL), policies are first safely trained offline using previously collected datasets and then further fine-tuned for tasks via limited online interactions. In a typical O2O-RL pipeline, candidate policies trained with offline RL are evaluated via either off-policy evaluation (OPE) or online evaluation (OE). The policy with the highest estimated value is then deployed and continually fine-tuned. However, this setup has two main issues. First, OPE can be unreliable, making it risky to deploy a policy based solely on those estimates, whereas OE may identify a viable policy with substantial online interaction, which could have been used for fine-tuning. Second--and more importantly--it is also often not possible to determine a priori whether a pretrained policy will improve with post-deployment fine-tuning, especially in non-stationary environments. As a result, procedures committing to a single deployed policy are impractical in many real-world settings. Moreover, a naive remedy that exhaustively fine-tunes all candidates would violate interaction budget constraints and is likewise infeasible. In this paper, we propose a novel adaptive approach for policy selection and fine-tuning under online interaction budgets in O2O-RL. Following the standard pipeline, we first train a set of candidate policies with different offline RL algorithms and hyperparameters; we then perform OPE to obtain initial performance estimates. We next adaptively select and fine-tune the policies based on their predicted performance via an upper-confidence-bound approach thereby making efficient use of online interactions. We demonstrate that our approach improves upon O2O-RL baselines with various benchmarks.