Agent 进展

Vision Language Models (VLMs) face significant challenges with ultra-long, interleaved image-text sequences due to the quadratic complexity of self-attention. Current solutions either resort to aggressive token pruning, risking irreversible information loss, or adopt efficient but less precise architectures, while largely ignoring the equally vital textual component. We introduce VLZip, a framework that unifies visual and textual compression for high-fidelity reasoning within a pure Transformer. At its core, VLZip hierarchically distills visual and textual segments into compact, layer-specific "soft prefixes" and injects them into each decoder layer's hidden states, drastically shortening the attention sequence while preserving fine-grained global context. To address deficient evaluations in the field, we also introduce LongVLBench, a new benchmark derived from video narratives that demands holistic, narrative-level reasoning. Extensive experiments show VLZip achieves leading performance on long-context multimodal reasoning, enabling training up to 120K tokens, a 6x increase over the baseline, and inference beyond 280K tokens with significantly reduced memory, while demonstrating the memory scalability to handle up to 2M tokens. By excelling at extreme context lengths where existing methods collapse, VLZip establishes an efficient and powerful new standard for long-context multimodal AI. Code is available at https://github.com/ShareLab-SII/VLZip.

Transformer-based large language models (LLMs) achieve strong performance across many tasks, but their Key-Value (KV) cache grows linearly with sequence length, creating a severe memory bottleneck for long-context inference. Existing heuristic eviction methods (e.g., H$_2$O and SnapKV) rely on static attention or positional signals that often fail to capture a token's future predictive influence. We propose DistillCache, a reinforcement learning framework that formulates KV-cache eviction as a sequential decision problem. DistillCache learns a lightweight policy network using rich internal model signals (attention statistics, value norms, entropy, and position) and trains it with REINFORCE via a per-step KL-divergence reward to preserve the full-cache output distribution. On a 7B-parameter instruction-tuned Transformer (Mistral-7B-Instruct-v0.3), DistillCache retains 94.2% of full-cache accuracy on LongBench at a 25% cache budget, outperforming both strong heuristic baselines (H$_2$O, SnapKV) by up to 2.7 absolute points and, under our re-implementations, concurrent RL-based methods (ForesightKV, RLKV) by up to 1.4 points on long-context tasks. On reasoning benchmarks, DistillCache is competitive with the best concurrent method and surpasses it under aggressive compression. It also delivers up to 2.1x full-cache throughput while maintaining competitive practical efficiency. These results highlight the effectiveness of learned, distribution-aware policies for memory-efficient long-context LLM inference.

Video Anomaly Detection (VAD) aims to identify anomalous events and localize their temporal intervals. Existing approaches exhibit a "when-what" dissociation: traditional DNN-based methods localize when anomalies occur but lack semantic understanding, whereas LLM-based methods explain what happens but neglect precise temporal grounding. We attribute this to the absence of a unified reasoning paradigm. Inspired by how humans inspect surveillance videos - glancing globally to form temporal hypotheses, scrutinizing suspicious segments, and thinking iteratively to correct errors - we study this global-to-local paradigm from two perspectives. We first propose Glance then Scrutinize (GtS), a training-free framework using static and dynamic textual guidance for coarse-to-fine anomaly grounding and understanding, balancing accuracy and speed. To break the ceiling imposed by frozen external modules, we further propose a tool-augmented agentic VAD method, where a multimodal large language model learns to invoke a video cropping tool, inspect densely resampled frames, and self-correct mislocalized hypotheses, via cold-start supervised fine-tuning followed by reinforcement learning with a joint answer-grounding reward. For training and evaluation, we extend our prior VAGU benchmark into VAGU-T (Video Anomaly Grounding, Understanding, and Thinking), comprising 7,567 real-world videos over 21 anomaly categories with human-validated grounding, explanations, QA pairs, and chain-of-thought tool-calling traces. We further introduce JeAUG, a metric jointly evaluating semantic interpretability and temporal precision. Experiments show that GtS substantially surpasses training-free baselines, while the agentic model delivers both higher accuracy and faster inference.

Aligning large language models (LLMs) with human values remains a major challenge, especially for trustworthy AI. While existing approaches such as RLHF, CAI, and their variants have achieved promising results, they often rely on a single-agent framework and a unified reward system. This limits their ability to capture ethical pluralism, adapt to diverse moral contexts, and reflect the dynamics of multi-agent moral reasoning. In this work, we propose a framework that utilizes multilayer combinatorial fusion for contextual value alignment (MCF-CVA). At the first layer of the framework, it instantiates multiple moral agents, each fine-tuned to represent a distinctive value. Their outputs are then expanded combinatorially using both score- and rank-combinations as well as average and weighted aggregations. These combined models are then reduced to the same number of initial moral agents. This expansion and reduction (EAR) process continues for multi-layers until a stopping criterion is reached. The MCF-CVA framework leverages cognitive diversity between agents to mitigate conflicts and redundancies across multiple agents, producing responses that better reflect contextual human values. The framework using the EAR algorithm is performed on the dual architecture of Euclidean score space and Kemeny rank space. Empirical evaluations demonstrated that the proposed framework outperforms single-agent baselines, multi-agent single-layer results, and previous aggregation approaches on standard metrics, showing that the MCF-CVA framework provides a robust and effective mechanism for advancing contextual value alignment in LLMs.

Agent Skills provide reusable capabilities to LLM agents. Agent Skill inconsistencies can expose undisclosed dangerous behavior or cause wrong Skill selection. Recent Agent Skill research has increasingly examined Agent Skill consistency detection. Existing methods evaluate behaviors or security-property graphs against predefined categories or declared scopes. More recently, PL-HCL uses an LLM-based model to learn consistency across metadata, instructions, and resources. However, declaration and implementation behavior can be mixed across text and code, and a concise declaration can correspond to multiple connected implementation steps. We present SkillConsist to address both challenges. An LLM separates declaration and implementation content into behavior records on the implementation and declaration sides, while static analysis supplements implementation records. These records form declaration and implementation behavior graphs, respectively. Starting from a behavior record on either side, bidirectional graph alignment searches the other graph for a candidate subgraph and expands it along behavior relations until it completely expresses the source-side behavior. Graph differencing identifies conflicts between aligned subgraphs and outputs the detection results. We construct a 633-Skill benchmark from ClawHub's 500 most-downloaded public Skills and 133 Skill-Inject packages. The benchmark contains 319 inconsistent and 314 consistent Skills and 442 localized inconsistency annotations. On this benchmark, SkillConsist achieves 86.85% precision, 89.03% recall, and 87.93% F1 for package-level detection, improving F1 over the best baseline by 20.43 percentage points. For localization, it achieves 67.60% precision, 58.14% recall, and 62.52% F1.

Translating a natural-language question into a SPARQL query that can be executed against a large knowledge graph requires resolving lexical ambiguity, grounding surface terms in the target ontology, and producing graph patterns that are both syntactically valid and semantically faithful. We present an agentic text-to-SPARQL system that goes one step beyond static tool-using agents: a researcher agent that, after each round of inference on a validation set, proposes and tests changes to its own prompts, rules, and tool-orchestration code. We instantiate the loop on DBpedia, evolve nine successive versions of the agent driven by a low-cost reasoning model, and deploy the best-performing configuration with two stronger backbone models. The study yields three observations: (i) self-improvement converges quickly and then achieves 0.22 overall accuracy on the 2025 DBpedia validation set; (ii) the bottleneck is consistently in basic-graph-pattern predicate selection, not in SPARQL syntax or modifiers; and (iii) several benchmark items appear to penalise correct queries due to property ambiguity in DBpedia, suggesting that future Text-to-SPARQL benchmarks should be scored using a combination of machine translation and information retrieval metrics.

Large language models perform strongly on medical knowledge benchmarks, but reliable clinical deployment requires agents to conduct defensible investigations over heterogeneous, longitudinal records: determining what evidence is needed, retrieving and reconciling structured and free-text data, grounding conclusions in verifiable evidence, and deferring cases that cannot be resolved reliably. We introduce CliniCARE-Bench (Clinical Calibrated Audit of Medical Reasoning in EHR), a benchmark for retrospective clinical audit: 25 clinician-validated scenarios instantiated as 750 patient-specific cases over real-patient-derived MIMIC-IV data. Systems investigate each case through a governed, logged tool environment for record retrieval, computation, and policy access, and return one of four verdicts---Yes, No, Indeterminate: Lack of Data, or Indeterminate: Medically Ambiguous---the last two separating missing evidence from residual medical ambiguity. Beyond verdict accuracy, we score patient-evidence and policy grounding, process adherence, calibrated abstention, reliability, and efficiency against case-level reference verdicts produced by independent multi-model adjudication and calibrated against Clinical Board review. Every retrieval, computation, and report is replayable, so the investigation trace is inspectable and scorable. To our knowledge, CliniCARE-Bench is the first deployment-oriented clinical-agent benchmark to jointly evaluate real longitudinal EHR investigation, claim-level evidence grounding, governing-policy use, process adherence, and calibrated abstention within a common patient-level adjudication framework. Across 16 agentic systems, four-way accuracy spans 65.3-76.1%, but raw accuracy overstates investigation quality. Defect-free accuracy, which credits a verdict only when correct and free of prohibited shortcuts, is 4.8-14.8 points lower and reorders the leaderboard.

Human-like cognition does not select past experience by topical similarity alone: affective significance and unresolved conflict also shape what becomes accessible. We present PsychoAgent, a cognitive architecture for LLM agents that separates factual and affective memory and integrates both through a conflict-aware executive controller. Affective memories are first filtered by semantic relevance and then re-ranked by salience, preserving topical fit while allowing emotionally important traces to enter the prompt. Across three controlled conflict scenarios, the full architecture retrieved more conflict-critical memories than semantic-affective and single-memory RAG baselines (0.933 vs. 0.500 and 0.667), with a small semantic-similarity cost. Five blinded raters evaluated 27 outputs. After within-rater standardization, the full architecture had the highest overall mean (+0.22 SD), but corrected pairwise differences were not significant. A three-day illustrative trace further shows persistent affect, offline memory recombination, and selective memory reweighting. The findings support affect-sensitive retrieval as an inspectable mechanism for modeling human-like conflict effects in LLM agents.

LLM agents increasingly adapt to recurring tasks by accumulating procedural knowledge in skills. These skills are lightweight, reusable textual artifacts that are loaded into the agent's context without weight updates. Recent methods refine skills through iterative task execution, failure diagnosis, and trajectory-guided text-space updates. However, existing frameworks lack explicit diagnosis--outcome feedback and treat deletion as a generic edit operation rather than a dedicated mechanism for consolidating accumulated knowledge. We introduce SkillProx, a proximal-gradient-inspired forward--backward framework that couples closed-loop diagnostic evolution with utility-aware proximal refinement. Motivated by a composite objective balancing task loss and skill complexity, the forward stage re-executes diagnosis-driven edits on the same task batch, rolls back regressions, and feeds measured outcomes into subsequent diagnoses. The backward stage decomposes the resulting skill into auditable knowledge units, estimates their contributions using a frozen leave-one-out utility audit, and applies validation-gated consolidation, demotion, or removal. Experiments on in-distribution and out-of-distribution benchmarks across multiple backbone LLMs show that SkillProx improves average accuracy by 3.0 percentage points over the strongest gradient-based baseline. Component ablations demonstrate the complementary effects of closed-loop diagnosis and proximal refinement.

Reliable hypothesis testing is the foundation of many empirical scientific claims. Large language model (LLM) agents are increasingly used to automate this process, as they can inspect datasets, generate code, and produce analyses end-to-end. However, we show that they frequently make subtle inferential errors that lead to incorrect conclusions despite correctly executed analyses. Existing benchmarks fail to capture this failure mode, as they rarely assess whether a reported p-value is statistically valid given the assumptions underlying the data. We address this gap by building P-Bench, a benchmark comprising 425 open-ended, realistic hypothesis-testing tasks spanning economics, biology, and medicine. Each task requires an agent to select a statistical method, compute a p-value, and draw a conclusion given only a scientific hypothesis and a dataset. We further introduce Fisher-R1, an open-weight LLM agent trained for rigorous hypothesis testing using synthetic tasks and reinforcement learning. On P-Bench, Fisher-R1-14B substantially improves over its backbone and outperforms strong proprietary and open-source baselines, including GPT-5.4 and DeepSeekV4-Pro, achieving a 21% average relative improvement in single-trial success over DeepSeek-V4-Pro, with gains up to 26% on the most challenging tasks. Our results demonstrate that current LLM agents lack reliable statistical reasoning for hypothesis testing and that reinforcement learning on tasks with verified statistical reward substantially improves reliability.

Large language models increasingly generate complete websites from natural-language descriptions, and reinforcement learning has become a central approach to closing their remaining functional gap. This training regime is bottlenecked by reward design. Hand-authored browser scripts are executable yet costly to write for open-ended requirements, while VLM and GUI-agent graders scale but may issue verdicts before observing the decisive state. We propose WebGrader, a self-evolving programmatic grader that autonomously derives the required interaction flows from each website request, represents each flow as an executable Flow Contract, and uses its execution outcome as an RL reward. WebGrader materializes the generated project in a live browser, grounds target actions against the source code and live DOM, and collects visual, DOM, response, and persistent-state evidence along the same browser trajectory. A residual-driven offline loop then discovers reusable verifier skills, screens them on disjoint validation pages, and freezes the promoted skill graph before policy training. By separating test planning, action grounding, evidence collection, and semantic judgment, WebGrader issues a Pass verdict only after observing the requested transition. On WebGen-Bench, WebGrader trains an 8B policy to a 52.01% functional success rate, outperforming a matched appearance-plus-script reward by 7.88 points and surpassing o4-mini and DeepSeek-v4-flash. On WG-core-250, the policy reaches a Full Score of 44.953 and surpasses Qwen3-Coder-480B.

Computer-use agents (CUAs) face a growing threat from indirect prompt injection, where adversarial instructions are planted in the environment such as web pages. In this paper, we introduce multi-step indirect prompt injection, a new attack class against CUAs in which the adversarial goal is decomposed into multiple innocuous-looking sub-steps and distributed across a chain of pages referenced along the agent's navigation path. We develop a pipeline to automatically decompose an adversarial goal under the constraint that the execution of the decomposed sub-steps must achieve the original goal while optimizing the innocuousness of each decomposed sub-step. With this pipeline, we build StepJack, a CUA safety benchmark with 480 test examples. On this benchmark, we evaluate six state-of-the-art CUAs and find that at a fixed decomposition depth, multi-step attacks raise attack success rate (ASR) on three of six CUAs, by up to 31.2 points (e.g., GPT-5.4-mini: 41.7% at single-step to 72.9% at three-step); averaged over the five CUAs that can reliably follow the reference chain (all but EvoCUA-32B), ASR rises from 31.3% at single-step to 36.9% at three-step. Dataset and code are available at https://github.com/BorealisAI/StepJack.

Software-Defined Vehicles (SDVs) expand the automotive attack surface across source code, runtime logs, and deployment topologies, while safety constraints forbid autonomous agents from acting without oversight. This paper presents CyberLLM, a multi-agent, LLM-orchestrated framework that autonomously detects vulnerabilities and executes remediations under a formal, runtime safety guard. Detection combines a deterministic layer (regex rules, AST analyzers, and topology graph checks) with an LLM refinement pass, so a high-recall floor is complemented by high-precision reasoning. A decision agent aggregates findings, tags them with a human-centric asset taxonomy, and selects a tiered response, ratcheting its confidence with signed cross-session memory and re-planning feedback. Every action is validated against four contextual security properties and an independent action-alignment oracle before it is allowed to run, and refused actions trigger escalation and re-planning. A symmetric attack pipeline generates and replays exploits so both sides can be exercised on the same scenarios. On an independent, ground-truthed benchmark of nine original automotive ECU modules in C, C++, and Rust seeding 47 layered vulnerabilities plus clean controls, the always-on deterministic layer covers 34% of the labeled vulnerabilities at perfect precision, and adding the grounded LLM refinement and completeness passes roughly doubles coverage to about 70% (F1 $0.83$) while producing zero false positives on the clean controls. The results indicate that LLM agents can perform useful autonomous cyber-defense when wrapped in a deterministic, auditable safety envelope.

The reasoning and agentic capabilities of large language models have expanded the range of applications they support, from short interactive exchanges to long, compute-heavy requests. LLM serving platforms today define response-latency service-level objectives, even though requests within the same service can differ by orders of magnitude in input length, generation length, execution cost, and the availability of reusable KV-cache state. As a result, requests governed by the same service level objective have different urgency: after accounting for the time required to execute them, some have substantial latency headroom while others have almost none. We define this headroom---the difference between a request's service level objective and its predicted remaining service time---as its per-request latency budget. We present Cascade, an LLM serving system that estimates and continuously updates this budget from request characteristics, KV-cache state, and current system load. Unlike prior SLO-aware schedulers that use deadlines to govern request ordering alone, Cascade uses a single per-request budget to jointly coordinate request scheduling and KV-cache management across the memory hierarchy. Its scheduler prioritizes requests with little remaining budget, while its memory manager uses the same budget to decide whether non-resident KV state should be restored or prefetched from a deeper tier, retained in HBM, or recomputed. By directing queueing and data-movement overhead toward requests that can absorb it, Cascade improves SLO-satisfied goodput while preserving fairness across heterogeneous request classes. On production traces across three large language models, Cascade improves goodput by up to2.4x and reduces SLO violations by 40% relative to the default vLLM first-come, first-served scheduler.

Deciding which of two agents is stronger means playing games until skill outweighs luck, and every game costs money, model inference, or expert time. Since the number of games needed is unknown, fixed-budget evaluations either keep paying after the result is settled or stop before the agents can be told apart, while naive optional stopping with an ordinary confidence interval invalidates the stated level. We make such an evaluation stop as soon as its evidence suffices, with the guarantee intact. The Action-Informed Value Assessment Tool (AIVAT) reduces variance in imperfect-information games through conditional mean-zero corrections, by a median $54\times$ across 15 LLM agent configurations spanning 71,439 paired Heads-Up No-Limit Hold'em (HUNL) hands, but does not say when to stop. We combine AIVAT with continuously monitored Confidence Sequences (CSs) into anytime-valid AIVAT (AV-AIVAT), whose online value model learns only from past games so that no game scores its own correction. At the nominal 95% level and a target precision of $\pm1$ Big Blind, raw outcomes need a median $74\times$ as many hands as AIVAT-corrected outcomes to stop under the Asymptotic CS (AsympCS). Exact finite-sample certification uses the Empirical-Bernstein CS (EB-CS), which needs an independently justified bound on corrected payoffs. We establish such a bound structurally for Leduc hold'em and characterize a width floor set by the CS's bet cap and that bound, which governs how much of a variance gain becomes earlier stopping; the descriptive HUNL EB-CS runs show a median $1.37\times$ stopping-time ratio. AV-AIVAT turns variance reduction into efficient, auditable early stopping while separating asymptotic screening from exact certification, so an evaluation can stop the moment its evidence suffices and hand a third party everything needed to recheck the verdict at that very stopping time.

As LLMs are increasingly deployed within agentic systems, their capabilities depend not only on the model weights but also on the harness: the prompts, tools, control flow, memory, and orchestration code surrounding them. This makes automated harness optimization — the iterative and evaluation-guided improvement of a harness by an AI system — both an important route to improving AI systems and a demanding capability for AI systems themselves. Yet the community lacks a common protocol for measuring how well frontier LLMs perform at this task. We introduce HarnessOpt-Bench, a benchmark for end-to-end harness optimization under expensive and stochastic evaluation. An optimizer, an LLM paired with a coding harness, receives a target agent's seed harness, graded evaluation feedback, and a fixed target-evaluation budget. It edits the harness and nominates a final candidate, which is scored by its normalized gain over the seed on a held-out test partition that remains inaccessible throughout search. A trusted execution environment enforces the evaluation boundary, meters target-agent resource use, and preserves candidate versions for audit. We evaluate 5 frontier LLMs as optimizers both under a shared coding harness and under their native harnesses across 4 downstream tasks, over 111 scored runs. Experiment results show that optimizer models separate more than the coding harnesses they act through, native harnesses are not consistently superior, and gains vary substantially across tasks and seed regimes. These results establish harness optimization as a measurable and discriminative capability with large space for improvement.

Large language models (LLMs) increasingly support complex professional tasks, yet their capabilities in rule-intensive document review remain insufficiently evaluated. National standard documents, such as China GB/T standards, offer a representative testbed: they are lengthy, highly structured, and governed by explicit rules for scope, terminology, normative wording, and cross-section consistency. Existing benchmarks focus on domain knowledge and question answering, largely overlooking intrinsic quality review for professional documents. Such reviews rely heavily on human experts, making them costly and difficult to scale. To bridge this gap, we introduce GB/T-Bench, the first benchmark for the structured review of national standard documents. Its GB/T Review Taxonomy is a hierarchical schema covering document structure, scope alignment, normative modality, terminology consistency, and normative references, with 25 diagnosable error types. A controllable counterexample generation mechanism combines deterministic rules and constrained LLM rewriting to process 488 documents into 7,306 traceable review error instances for evaluation. We also develop a diagnosis-oriented evaluation protocol requiring exact matches on error location, review dimension, and error type, plus document-level coverage metrics. We further propose GB/T-Reviewer, a multi-agent framework that converts review knowledge into specialized skills and coordinates global inspection, targeted diagnosis, rule scanning, and result verification. Experiments with 14 mainstream LLMs reveal a substantial human-LLM gap: the strongest model achieves only 0.3280 CMCS versus 0.6640 for experts. GB/T-Reviewer raises the best CMCS to 0.5094, showing the value of structured skill coordination for rule-intensive document review. This work paves the way for trustworthy AI in standardization and other high-stakes document domains.

Tool use transforms LLMs into agents that act beyond their training data, and for code-capable models, programmatic tool calling extends this further by replacing rigid JSON calls with scripts that chain and parallelize naturally. However, a systematic evaluation of tools as code on an established benchmark across current and prior model generations under real-world task conditions has not been conducted. In this work, we empirically compare programmatic tool calling (PTC) to native JSON tool calling across 14 language models on BFCL v4. In the programmatic tool calling paradigm, tools are exposed as typed Python stubs that the model invokes through code, with execution and results handled in a single agent turn. Programmatic tool calling matches or exceeds native JSON tool calling in 11 of 14 models on BFCL v4, with the GPT-5.6 family achieving a 10.6% improvement over the JSON tool calling baseline. Further, it matches or outperforms baseline in 13 of 14 models under parallel fan-out, and holds stable under context rot conditions where baseline degrades 2.3% on average. Our results demonstrate that programmatic tool calling is a viable and robust alternative to JSON tool calling, with performance tracking model capability across release generations.

LLM-based agentic systems have shown remarkable capabilities in complex domains, while suffering from cascading errors and difficulty in debugging. Critical error detection aims to locate the earliest error step in a failed trajectory that is responsible for the final failure. However, progress faces two main challenges. First, long trajectories make it difficult to identify individual errors, since the evidence for judging a step may be scattered across distant instructions, observations, and prior context. Second, failed trajectories often contain multiple local errors with different downstream effects, only some of which remain responsible for the final failure. In this work, we propose TrajDebug, an error-lifecycle tracing framework that addresses long-trajectory error discovery with multi-granularity history compression and evidence-based error identification, and supports critical attribution by tracing each error's resolution status and terminal impact. We further construct TrajErrBench, a benchmark of 486 manually annotated failed trajectories from Tau2Bench and SWE-Bench Pro, covering realistic tool-use and coding scenarios. Experiments across diverse agent benchmarks show that TrajDebug achieves the best overall performance over existing baselines, and application studies further demonstrate that its diagnoses provide actionable feedback for improving downstream agent success. We will release the codes and data to facilitate further research.

Context-aware recommender systems have long recognized that factors such as location, time, and weather shape where and what people choose to eat. Existing weather-aware food and point-of-interest recommenders, however, typically treat weather generically — mapping conditions to preferences through hand-crafted rules or specially trained context models — and do not capture that the culturally appropriate response to weather is itself region-specific: a rainy evening calls for hot tea and fried snacks in one culinary culture and for very different comfort food in another. Encoding such weather-by-region-by-cuisine interactions as explicit rules or training data is brittle and does not scale. We present a weather- and location-aware agentic dining-recommendation system that takes a different approach: a large language model (LLM) orchestrates tools for location and weather retrieval and then reasons in natural language over the combined context, drawing on the cultural and culinary world knowledge already latent in the model to produce region-sensitive, weather-appropriate recommendations without per-region rule tables or specialized training. We describe the agent architecture, the tool-orchestration flow (Google location services and a weather service feeding an OpenAI LLM), and the reasoning mechanism, and we report on a working prototype that was implemented and briefly deployed end-to-end. We discuss design trade-offs — cost, latency, ambiguity handling, and fallbacks — and we are explicit about limitations, including the absence of a formal user study and the risk of cultural stereotyping in locality-based inference. The contribution is architectural: a simple, extensible pattern for incorporating environmental and cultural context into agentic recommendation through LLM reasoning rather than engineered rules.

Clinical machine learning (ML) has the potential to support high-stakes medical decision-making, but reliable deployment is often constrained by scarce, heterogeneous, and temporal complexity. Developing effective ML pipelines for such data remains time-consuming and error-prone, while existing automated machine learning (AutoML) systems only partially address this challenge because they largely rely on brute-force search over predefined spaces and lack explicit reasoning and memory. We therefore reformulate AutoML for small clinical data from exhaustive search to reasoning-driven refinement. We propose DoctorAgents, an agentic AI framework that autonomously constructs and optimizes end-to-end ML pipelines through specialized large language model (LLM) agents for generation, validation, and refinement. DoctorAgents backpropagates natural-language feedback through textual gradient descent to perform targeted updates without exhaustive search. Experiments across diverse clinical tasks show that DoctorAgents consistently outperforms established AutoML baselines while producing more interpretable task-specific representations.

Multi-agent orchestration frameworks are moving from demos to production, yet benchmarks typically report task accuracy without diagnosing why a pipeline failed, where a cascade began, or which routing decision caused the breakdown. OrchestraBench evaluates failure, recovery, and decomposition through a controlled, seed-reproducible failure-injection harness over templated enterprise workflows. It introduces cascade radius and per-failure-mode recovery as primary metrics and compares routing policies with bootstrap confidence intervals and paired tests. On a 26-case gold-labelled diagnostic, a keyword/flag router scored 0% on adversarial cases with misleading or missing surface flags, whereas an intent-reasoning model router scored 100%, matching the oracle. Controlled mechanism probes with a real Claude agent over a verifiable arithmetic dependency chain revealed three failure-handling tiers across five MAST modes: tool faults recovered fully (1.0), ambiguous delegation recovered partially (0.30), and three latent or semantic modes never recovered (0.0). This ordering persisted when the computation was reframed as a loan-approval workflow and across Sonnet, Opus, and Haiku, although absolute rates shifted with context. Blind retry reproduced latent faults and increased time to detection, indicating that detection and attribution are necessary for containment. Cascade radius increased with pipeline depth (mean 0.9 to 4.7 across depths 3-7). A trusted-state repair ablation showed that apparent containment gains primarily came from the trusted-state signal rather than autonomous detection. These results are controlled-chain mechanism probes, not domain-workload claims.

High-quality annotation of artworks is essential for computational art research, yet extracting implicit semantics remains challenging due to the reliance on culturally grounded meanings and deep contextual knowledge behind the images. Current AI-assisted annotation tools often lack assistance or rely on one-way workflows where experts have to perform extra manual calibrations to improve AI models, resulting in limited efficiency. To address this, we propose Bidirectional Human-AI Augmentation(BiHAA), a closed-loop framework in which skills and domain knowledge base evolve through real-time interaction and bidirectional HAI augmentation. Informed by a formative study with 20 artwork annotators from different backgrounds, we implement this framework in ArtAnno, an artwork annotation system driven by a multi-agent architecture. The system includes a Proactive Agentic Support Module, where AI augments humans through semantic mining and label suggestion, and an Interaction-Driven Evolution Module, where human expertise continuously enhances the AI through distilling annotation trajectories into reusable experience. Evaluation through a user study and two case studies demonstrates that our framework and system improve annotation efficiency, enable knowledge accumulation, and reduce the effort of information seeking and verification for annotators with limited domain expertise. We conclude by discussing broader implications and future directions.

Long-horizon reasoning requires an agentic runtime that can persist when evidence supports its current approach and pivot when measurements reveal failure, hidden constraints, or a misspecified objective. We present Argus, a persistent, self-evolving runtime in which Manager, Planner, Engineer, and Reviewer execute bounded missions over durable project state. Argus separates stable user intent from operational objectives, constraints, and verification criteria, and admits memories, skills, procedures, verifiers, routing decisions, and rejected routes only after role-owned review and, when available, task-native verification. Model weights remain fixed; self-evolution occurs through persistent runtime state and control policy, with autonomous execution between operator-owned escalation points. Across seven GPT-5.5 benchmark arenas, Argus achieves about 78% on SWE-Bench Pro versus 59% for Direct Copilot while using 1.41 times the aggregate tokens. After verification-gated self-evolution, mature SWE-Bench waves use 21% fewer solve-input tokens and 15% less active workflow time per task than startup waves, while recording 34 verifier recoveries and 22 strict review-loop rescues. Argus also reaches 76.8% on AARRI-Bench and a 28.0-point gap on mathematical data synthesis, with competitive GPU-kernel and language-model-training results. Beyond benchmarks, an optimized RWKV6 kernel was merged upstream; a multi-day mathematics campaign retained falsified routes and proof-backed frontier updates; and six paper pipelines completed 254 missions with 16 stage rollbacks. These results show that a fixed-weight, self-evolving harness can revise, recover, and accumulate verified approaches while producing structured trajectories for future supervised and reinforcement learning.

AI-supported care planning can help clinicians, patients, caregivers, and care teams coordinate complex decisions across clinical, functional, psychosocial, and environmental needs. However, many AI systems present recommendations as fixed outputs, limiting stakeholders' ability to inspect, challenge, and revise plans when they conflict with clinical judgment, patient values, or real-world feasibility. We present CoPlan - a Co-Intelligent and Contestable Interface for Human-AI Care Planning. CoPlan uses a multi-agent workflow in which specialized AI agents generate candidate interventions and supporting or challenging arguments, while human care planners can accept, reject, modify, or add arguments before final plan generation. Through this design, CoPlan combines co-intelligence, in which humans and AI agents contribute complementary expertise, with contestability, where recommendations remain open to inspection, revision, and justification. We demonstrate CoPlan in an aging-in-place care planning scenario. The system supports adaptive care team recruitment, role-based argument review, final care plan generation, and practical follow-up through scheduling agents. This work contributes a contestable care planning interface and a design framing for trustworthy human-AI care planning that preserves human agency and clinical accountability.

Agents for long term reasoning require a memory that can be efficiently and effectively updated over time, as new facts and external feedback continue to arrive. Recently, graph memory has been adopted to offer structural organization for multi-hop retrieval and reasoning. However, existing methods store all memories in a flat graph, and accumulated historical memories can introduce irrelevant contexts and increase the cost of evidence selection during retrieval. Moreover, they typically update memory units independently, requiring repeated unit-wise rewrite to cover related changes. To address these issues, we propose HiGram, an evolving hierarchical graph memory framework with path-level localization and rewriting. Specifically, we first propose a hierarchical graph memory, which organizes the memory into coarse-to-fine architecture composed of upper-level nodes and MemoryUnits, thereby reducing the amount of irrelevant information during retrieval. We further propose MicroGraph-based path-level localization, which leverages query and update conditioned MicroGraphs to identify support subgraph and evidence path before rewrite. Finally, we propose a coordinated rewriting method that jointly revises intra-unit memory and inter-unit dependencies, enable valid dependency structures updating in the localized evidence path. Experiments on benchmarks for long-term conversational question answering and conflict-aware memory evaluation demonstrate that our method demonstrate substantial improvements over baselines in answer quality and token efficiency. Besides, our method improves answer accuracy and query-valid evidence selection under dynamic, static, and conditional conflicts.

Long context reasoning in large language models (LLMs) is usually constrained by the fact that a single inference trajectory has to simultaneously explore the context, store intermediate state, verify evidence, and produce the final answer. This becomes particularly difficult in tasks that require extraction, counting, ordering, or multi-hop reasoning, where an early mistake can propagate until the final response. In this work, we propose Chained Recursive Language Models (Chained RLM), an inference-time architecture, in which the same underlying model is called repeatedly as a sequence of fresh reasoning roots. Each root receives the original problem and context, but does not inherit the full conversational history. Instead, it receives a compact plain-text summary, a plain-text blackboard, and some durable task-specific artifacts written by predecessor roots. The motivation is to manage the context by chopping into partial tasks rather than one large inference response; in each staged computation, intermediate artifacts can be inspected, corrected, and extended by a later fresh inference by the same model. We describe the system model, handoff mechanism, artifact workspace, and evaluation protocol for this system. We study when fresh-context artifact continuation gives a measurable gain in accuracy over direct LLM answering even with recursive tool-calling.

Prompt injection remains a critical threat to LLM agents, yet existing defenses treat each task as a self-contained problem, independent of previous encounters. In practice, user requests are often underspecified: they describe the desired outcome without fully specifying acceptable behavior. An injection can exploit this ambiguity, causing the agent to complete the task in a way the user would reject. As the user's expectations become clearer through concrete cases, a defense should learn from each encounter and apply what it learns to the next. Inspired by adaptive immunity, we propose AgentAntibody, which equips LLM agents with a self-evolving immune system against prompt injection. AgentAntibody represents its evolving understanding of the user's security boundary as a persistent library of antibodies. At runtime, the library recognizes threats to this boundary and mounts corresponding immune responses. Across encounters, it evolves to strengthen the agent's immunity to future attacks. Extensive experiments across three benchmarks and four backbone LLMs show that, by learning the user's boundary through experience, AgentAntibody outperforms existing defenses in preventing harmful actions while preserving legitimate task completion, even when the harmful and legitimate actions are both compatible with the stated task.

Open-World Learning (OWL) pipelines for oil well anomaly detection have recently been shown to combine autoencoder-based detection, multiclass classification, and Mahalanobis-based novelty detection on the public 3W dataset. These pipelines answer what happened, but they do not explain why the model believes it or what the operator should do next, and they do not put a human-readable name on the novelty clusters they discover. This paper evaluates a Large Language Model (LLM) agent layer placed downstream of the OWL pipeline, designed as a companion to the published upstream methods rather than a replacement. Using the Qwen3.5-397B-A17B Mixture-of-Experts model served via NVIDIA NIM, the agent receives structured sensor metrics and upstream classification or novelty assertions, and returns natural-language justifications, confidence-ranked critiques, and consolidated names for detected novelties. Across three studies spanning 989 real well-file segments from the 3W dataset, the agent achieved $35.1%$ top-1 / $63.9%$ top-3 (95% CI [56.9, 70.4]) classification on all nine classes, $71.7%$ top-2 validation [64.8, 77.6] with precision $0.91$ [0.84, 0.95] across 7 probed classes, and $89.7%$ novelty detection [87.0, 91.9] with stable cluster naming on 5 of 7 hidden classes. The agent is not a standalone classifier. Its role is to: (1) confirm upstream decisions when sensor evidence supports them, (2) justify decisions in sensor-grounded language operators can audit, (3) flag disagreement when upstream labels are implausible, and (4) name novelties so that clustered unlabeled events arrive at the engineer with a consolidated human-readable label. The goal is to close the explainability gap that currently blocks deployment of OWL pipelines in operational settings.

High-resolution images and long videos provide vision-language models with rich context for multimodal reasoning and fine-grained perception, but the resulting long visual token sequences make large language model-side computation and memory costly. Existing visual token reducers often operate at prescribed rates, while recent methods adapt token counts across inputs using method-specific learned thresholds or importance predictors. We introduce RUTA, a principled Rate-Utility Token Allocation method that performs pre-LLM reduction by jointly learning which tokens to retain and how many to allocate to each image-query pair. RUTA constructs query-conditioned candidate tokens and predicts a retention probability for each candidate. During training, these probabilities parameterize independent Bernoulli gates, while their sum provides a differentiable training-time estimate of the token count for each pair. Retained tokens serve as anchors that aggregate information from non-retained tokens according to semantic affinity and spatial proximity. RUTA is optimized with a penalized rate-utility objective that balances downstream task loss against expected token usage. Averaged across five benchmarks and measured relative to each backbone's full-token baseline, RUTA uses only $2.0%$ and $4.2%$ of visual tokens while preserving $88.2%$ and $94.4%$ of task performance on LLaVA-NeXT-7B and Qwen3-VL-8B, respectively.