Artificial Intelligence · Cross-listed from cs.CR
SkillDRE: Dual-Stage Red-Team Evolution of Agent Skills via Pre-Execution and Runtime Feedback
Pengyu Zhu, Jingyi Yang, Yi Liu et al. (+2 more)
Agent skills package instructions, executable code, and task-specific resources into reusable artifacts that
agents can improve using execution feedback. The same mechanism also enables attackers to evolve malicious skills, making them more effective and less detectable. However, a candidate skill may pass pre-execution scanning yet fail to realize its target under runtime defenses, while a revision that repairs execution may introduce new scanner findings. We introduce SkillDRE, a fully automated framework for evolving complete malicious skill packages through a dual-stage feedback loop. Given a benign task and its associated skills, SkillDRE autonomously constructs and validates a task-conditioned malicious objective and a verifiable judge rule. It then holds both fixed while evolving the skill implementation, with preservation of legitimate task capability. SkillDRE combines scanner-guided
evolution with runtime-guided refinement informed by execution outcomes observed under runtime defense. Each runtime-guided revision returns to the pre-execution stage for rescanning and further optimization before re-execution, forming a cross-stage closed loop. Evaluated on SkillsBench across four victim models, SkillDRE achieves an average attack success rate of 45.28%, exceeding the strongest baseline by 40.3%, while its final submitted skills receive no SkillScan findings and largely preserve benign-task performance. These results show that two-stage defense feedback can serve as a useful learning signal for adaptive red teaming and that evaluating either defense stage in isolation can miss the resulting attack capability. Codes is available at https://github.com/whfeLingYu/SkillDRE
Read the full abstract ▾
All authors Pengyu Zhu, Jingyi Yang, Yi Liu, Li Sun, Sen Su
Artificial Intelligence
Function Over Form: Distributional Orthogonalization in Mixture-of-Experts with Replica Expert Mechanism
Jinfan He, Yunzhuo Liu, Kai Zhang et al. (+3 more)
The scaling of LLMs increasingly relies on MoE architectures to decouple active computation from total parameter count. However, the efficacy of MoE is often constrained by expert collapse and representation redundancy, both leading to underutilization of model capacity. To address these challenges, this paper proposes Distributional Orthogonalization Loss (DO-loss), an auxiliary regularization that shifts the focus from static weight diversity to dynamic routing behavior. By representing each expert's
token assignment history as a high-dimensional binary load signature, DO-loss penalizes signature overlap to prevent expert collapse while encouraging functional specialization. To align this algorithmic design with system efficiency, we further introduce the Replica Expert Mechanism (REM), which improves load balancing through a two-tiered strategy: adjusting replica expert placement at the global-batch level and performing real-time
token dispatching at the micro-batch level. Empirical evaluations demonstrate that our method outperforms the evaluated routing
algorithms on downstream tasks for both 4.8BA0.5B and 30BA3B MoE models, while maintaining comparable training efficiency.
Read the full abstract ▾
All authors Jinfan He, Yunzhuo Liu, Kai Zhang, Weidong Han, Key, Rayying
Artificial Intelligence
Memory as a cache: Exact context reuse and deletion by construction
Shengyao Wang and Jiang Liu
The KV cache of a
transformer entangles every
token's representation with its entire prefix: a passage encoded once cannot be reused under a different prefix or removed without recomputing everything after it, so exact cache reuse is limited to shared prefixes. We present SMem, an architecture whose context representation is a cache by construction. A block-local encoder maps each block to
memory rows independently of other blocks, and a reader conditions generation on their union through cross-
attention. For every parameter setting,
memory composes exactly at fixed block indices, deleting a block is an exact $O(b)$ update for $b$-
token blocks, and the
memory state is independent of the edit path. At $4\times$ the training context, under the shared recipe, SMem retrieves planted needles beyond any trained-length window (exact match 0.14-0.28 at distances of 31 and 63 blocks), where learned-position, RoPE, and Block-
Attention-style
transformers all score at most 0.02. A fully cached context is served by computing one block alone at a near-constant 3.1-6.2 ms, whereas cold prefill grows with context; batched decode stores 34-38% fewer KV rows and runs 1.4-1.7$\times$ faster when bandwidth-bound; and deletion beats suffix recomputation by 8.5$\times$ at 512 blocks and 452$\times$ at 4096 blocks (32-256$\times$ the trained length, probing the cost model rather than a served regime). The cost is a perplexity gap of -4.7% to +2.8% (negative favors SMem) against a parameter-matched
transformer with the same positional scheme, at 160M-1.5B on FineWeb-Edu across two recipes and a learning-rate search. SMem also composes with RoPE: at 160M and 410M the composite matches or leads the matched
transformer and closes 29-59% of SMem's gap to a RoPE
transformer. Dropping prefix
entanglement thus keeps perplexity comparable while making the cache exactly composable and editable.
Read the full abstract ▾
Artificial Intelligence
Beyond Scripted Search: Sample-Efficient Reward Discovery via Agentic Black-box Optimization
Minghao Li, Rui Tan and Ruihang Wang
Designing dense reward functions for low-level reinforcement learning (RL) control remains difficult. Recent work uses large
language models (LLMs) to iteratively generate and refine reward functions using policy-training feedback within scripted search
algorithms. However, evaluating each candidate requires a full RL training run, making sample efficiency a central challenge for reward search on complex control tasks. To address this limitation, we propose an Agentic Reward Black-box Optimization (ARBO) framework, in which an LLM
agent builds the search strategy at run time from an evaluation history maintained as its persistent workspace. The evaluation history comprises two components: observations maintained by the evaluation oracle, including candidate scores, per-term training curves, and error tracebacks; and an
agent-maintained belief that records diagnoses and intended next steps. The
agent queries both with tools and generates the next batch of reward candidates, rather than generating them in a single pass from a fixed
prompt. Across four control domains, ARBO achieves gains of 29.9% in manipulation success rate and 192.8% in power-grid score over baseline means under a shared evaluation budget. Ablations examine each component's contribution and sensitivity to backbone choice.
Read the full abstract ▾
Artificial Intelligence
SCLATE: a Substrate for Continual-Learning Agent Training and Evaluation
Youngmok Jung, Sirajul Salekin, Henry Tran et al. (+3 more)
Continual-learning
agents are systems of models, harnesses, and
memory operating over long multi-session horizons. Evaluating and training them requires interleaving tasks with
agent-side events such as session stop and start, crons, and
memory consolidation. Yet existing
benchmarks and training frameworks schedule only the
benchmark's own events, leaving each
benchmark and
agent pair to build a custom scheduling loop. We present SCLATE, an execution substrate where
benchmarks and unmodified
agents each add their events to one open event scheduler through an adapter. A hybrid simulated clock runs these events on a shared timeline, flowing in real time while the
agent works and skipping idle gaps, which compresses a month-long scenario into hours. SCLATE also serves as a rollout engine that runs any
agent's harness and
memory unmodified, recording the
tokens and log probabilities of every model call through an in-container proxy. We port seven
benchmarks to SCLATE and compare ten unmodified harness and
memory configurations head to head on ten models. The comparison shows that an added
memory system does not reliably beat the harness's native
memory and that models differ widely in how they use the same harness and
memory. We then post-train Qwen3.5-4B through unmodified harnesses and
memory systems. The model learns to use both, reading 6.8x fewer file lines with a 16.7-point higher SWE-bench Verified pass rate, and writing richer
memory records, while its held-out MetaClaw accuracy rises by up to 11.8 points.
Read the full abstract ▾
All authors Youngmok Jung, Sirajul Salekin, Henry Tran, Javier Movellan, Zhao Huang, Manjot Bilkhu
Artificial Intelligence
Reward Hacking and Agent Containment Failure: A Monte Carlo Study Based on the 2026 Hugging Face Incident
Murat Ozer, Bulent Erenay and Ibrahim Berber
The July 2026 intrusion into Hugging Face production infrastructure showed how reward hacking can become an external cybersecurity incident when a capable
agent encounters weak containment boundaries. This study develops a probabilistic risk model linking five stages: reward hacking, containment escape, usable access, persistence, and failure of detection. A Monte Carlo simulation evaluates 100,000 runs under each of four control configurations. Input distributions represent explicit uncertainty and are used for comparative analysis rather than real-world frequency prediction. Under the stated assumptions, layered controls reduce simulated external-incident probability substantially more than network isolation or monitoring used alone, an ordering that holds under independent plus/minus 25% perturbation of every coefficient in the model across 300 draws. Sensitivity analysis shows that
agent capability and weaknesses in monitoring, authorization, and credential control exert the greatest influence on modeled risk. Human temporal discounting and metric gaming provide a behavioral analogy for short-horizon optimization, but the study does not infer that AI
agents experience gratification or human motivation. The results support treating cyber-capable
agent evaluations as hostile security zones in which indirect egress, shared infrastructure, credentials, and evaluation artifacts must remain outside the
agent's effective authority.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.HC
DashAct: A Progressive Diagnostic Benchmark for GUI Agents in Interactive Dashboard Analysis
Chuhan Zhang, Qi Xie, Ziyue Wang et al. (+4 more)
Interactive dashboards require users to reveal and connect evidence across stateful interactions. Although graphical user interface (GUI)
agents could automate this process, existing dashboard
benchmarks primarily report final answers or task success. They provide limited insight into whether failures arise from maintaining the analytical process, selecting actions, or grounding visual targets. We introduce DashAct, a progressive diagnostic
benchmark that evaluates the same annotated dashboard interactions under systematically varied contextual and spatial support. DashAct contains 348 human-verified interaction trajectories with milestone dependencies and hierarchical target annotations. Its three stages evaluate end-to-end execution, verified-context action prediction, and target-conditioned grounding under four visual conditions. Aligned action and spatial annotations connect outcomes across these settings, providing diagnostic evidence for interpreting online failures. Experiments reveal persistent difficulties in action prediction and fine-grained grounding, alongside model-dependent responses to contextual and spatial support. A Qwen-based adaptation study further improves target localization and online Milestone Coverage, illustrating how DashAct can guide practical
agent improvements.
Read the full abstract ▾
All authors Chuhan Zhang, Qi Xie, Ziyue Wang, Jianing Yin, Yunfan Zhou, Dazhen Deng, Yingcai Wu
Artificial Intelligence · Cross-listed from cs.LG
TimeES: Probabilistic and Deterministic Time Series Forecasting via Evolutionary Spectra
Weiwei Ye, Renhe Jiang, Hangchen Liu et al. (+2 more)
Real-world time series are inherently non-stationary, with trends, periodic patterns, and uncertainty evolving over time. While the
Fourier domain offers a natural lens to model time series, current deep learning approaches do not explicitly model
evolution and randomness in the
Fourier spectra, which limits their ability to accurately predict both the expected trajectory and its uncertainty in non-stationary time series. Motivated by Evolutionary Spectra (ES) theory, we propose TimeES, a general framework that enables probabilistic and deterministic
forecasting via the evolutionary spectra theory. Specifically, we derive a parameterizable evolutionary spectra formulation, recasting non-stationary random process modeling as learning an evolving representation modulated by random variables. Furthermore, we reduce the complexity of the estimated spectra from O(NM) to O(NK), where K << M/2, by exploiting Hermitian
symmetry and spectral energy sparsity for frequency selection. Based on a simple linear backbone, our proposed TimeES achieves consistent state-of-the-art performance across both deterministic and probabilistic
forecasting tasks, with high efficiency and interpretability. Code is available at: https://github.com/wwy155/TimeES.
Read the full abstract ▾
All authors Weiwei Ye, Renhe Jiang, Hangchen Liu, Dongyuan Li, Yoshihide Sekimoto
Artificial Intelligence · Cross-listed from cs.LG
Measurement Boundaries in LLM Financial Agent Evaluation: Fixed-Tape Execution and Multi-Defect Auditing
Weicheng Xue
What controls are needed to interpret execution performance and audit scores in LLM
agent evaluations? We study two limits on these interpretations in a financial
agent harness. In Study~A, comparing independent runs under idealized and stressed execution on three synthetic settings that share one 24-day upward phase mixes the execution rule with fresh model responses and portfolio feedback: the parsed decision paths agree in only $19.8\%$ of $450$ pairs. Replaying each stored response tape through both execution destinations gives a narrower result. Conditional on those responses, stressed execution changes total return by $-0.0170$ (95% interval $[-0.0230,-0.0117]$), or $10.4\%$ of the idealized baseline, and ten seed clusters do not resolve the model ranking. Study~B corrects an incomplete answer key and replaces legacy tasks with matched zero-, one-, and two-defect tasks under an explicit multi-label
prompt. The drop in target violation recall from one to two defects is positive in five of six combinations of auditor and source (median $0.267$), with three surviving Holm correction. Yet the auditor that includes both target labels most often has micro-precision $0.149$, emits findings on $98/100$ zero-defect tasks, and returns the exact dual-defect set in only $21/100$ cases. Target recall by itself therefore gives a poor account of audit quality on this construction. The studies address different limits: what an execution comparison estimates, and what target recall captures. Together, they show how fixed conditions and diagnostic controls bound the claims a score can support.
Read the full abstract ▾
Artificial Intelligence
AuthorityLens: Rethinking LLM-Based Agent Systems Through the Lens of Authority
Shaojin Chen, Huihao Jing, Wun Yu Chan et al. (+7 more)
LLM-based
agents are increasingly deployed with authority over consequential resources and decisions in real systems. These
agents often operate alongside human and LLM-based participants who hold different forms of authority. Yet workflow roles, permission settings, and review mechanisms do not necessarily reflect the authority realized in practice. We introduce AuthorityLens, a framework for measuring a system's authority structure. Starting from an authority portfolio, we evaluate a system along three dimensions: what the system is authorized to do (System Authority), how much joint participation is required to exercise that authority (Authority Separation), and how much authority each participant holds (Principal Authority). We derive these measurements from the minimal combinations of participants sufficient to realize each outcome across admissible runtime states. We apply AuthorityLens to Codex, OpenCode, and Gemini CLI across 13 operating configurations over a common portfolio of
agent operations. We find that nominal configurations do not map cleanly onto realized authority. In Codex, Full Access changes System Authority only marginally while substantially concentrating authority in the executing Assistant. OpenCode's Build and Plan configurations have the same System Authority and Authority Separation despite different workflows and root-level permissions. In Gemini CLI, model-based review increases Authority Separation without changing System Authority. Principal Authority further distinguishes authority replication from authority separation: spawned or delegated
agents can become alternative holders of the same authority without increasing the required joint participation. Together, these results demonstrate that AuthorityLens provides a unified framework for measuring and comparing realized authority structures across
agent systems.
Read the full abstract ▾
All authors Shaojin Chen, Huihao Jing, Wun Yu Chan, Wenbin Hu, Jiaxing Li, Wu Pandy Pui Ching, Kshitij Bhatia, Xinlei He, Haoran Li, Yangqiu Song
Artificial Intelligence · Cross-listed from cs.LG
DiffPTS: Rethinking Diffusion ELBO for Probabilistic Time Series Forecasting
Weiwei Ye, Dongyuan Li, Hangchen Liu et al. (+3 more)
Probabilistic time series
forecasting requires modeling and predicting complex and time-varying distributions. Recently, Denoising
Diffusion Probabilistic Model (DDPM)-based approaches have shown promise by equipping the dif- fusion process with pretrained mean and variance
estimators to accommodate distributional shift. However, these methods typically follow the standard DDPM framework and consider only partial components of the evidence lower bound (ELBO), treating the training of
estimators as designed regression tasks separate from the variational inference framework. To address this, we rethink the ELBO under the Location-Scale Noise Model (LSNM) and find that it naturally induces a Gaussian negative log likelihood objective for the
estimators and inherently defines a joint training objective that unifies recent
diffusion paradigms for probabilistic
forecasting. Building on this principled ELBO reformulation, we propose Diff- PTS, a general framework that enables end-to-end optimization of all components within the ELBO. Across multiple
benchmarks, DiffPTS consistently outperforms recent models, achieving state-of-the-art performance with an average CRPS/MSE reduction of over 14.53%/16.55% compared to existing
diffusion-based methods. The code is available at https://github.com/wwy155/DiffPTS.
Read the full abstract ▾
All authors Weiwei Ye, Dongyuan Li, Hangchen Liu, Haotong Jiang, Yoshihide Sekimoto, Renhe Jiang
Artificial Intelligence · Cross-listed from cs.LG
Black-Box Auditing of Epistemic Reliability in Multi-Agent Debate Distillation
Derui Wang, Zewei Shi, Rayne Holland et al. (+4 more)
Debate
distillation adapts weaker verifiers using multi-
agent debate transcripts to improve their judgement in subsequent debates, but gains on monitored tasks do not establish reliability on related unmonitored tasks. We study epistemic reliability degradation, in which adaptation preserves monitored performance while reducing support for correct responses on hidden tasks. We consider an
adversarial debater that manipulates debate arguments while defending the correct monitored response, and ask whether the resulting degradation merely reflects catastrophic forgetting and whether standard evaluation can detect it. To address these questions, we propose ER-Audit, a two-stage black-box
auditing framework that compares frozen verifier checkpoints before and after adaptation, and introduce two evaluation
benchmarks pairing monitored and hidden task
prompts grounded in shared contexts. ER-Audit searches for counterexamples to non-degradation by evaluating semantically valid paraphrases and, if none is found, uses independent paraphrases for sequential hypothesis testing. We derive anytime-valid lower confidence bounds on the non-degradation probability, allowing data-dependent stopping within a finite budget. We further establish a common lower bound across fixed paraphrase distributions and extend it to distributions within a bounded total variation distance of their mixtures. Our experiments show that higher hidden-task accuracy can coexist with more counterexamples to non-degradation and lower non-degradation bounds. This divergence challenges explanations based solely on broad catastrophic forgetting and shows that
auditing can uncover selective hidden-task degradation concealed by aggregate performance gains. Our code and
benchmarks are available at https://github.com/CSIRO-CQS-AI-
alignment-Team/Epistemic-Reliability-Auditor.
Read the full abstract ▾
All authors Derui Wang, Zewei Shi, Rayne Holland, Ruoxi Sun, Xingliang Yuan, Jason Xue, Liming Zhu
Artificial Intelligence · Cross-listed from cs.CV
EyeVQA: Benchmarking Ophthalmic Vision-Language Models from Recognition to Spatial Grounding
Gujie Shao, Zixun Xie, Xuechun Xing et al. (+7 more)
Vision-
language models (VLMs) have shown increasing potential for medical image understanding, yet their capabilities in ophthalmic imaging remain insufficiently characterized. Existing ophthalmic
datasets are typically designed for individual diseases or specialized tasks, making it difficult to systematically evaluate whether VLMs can move beyond disease recognition toward comparative
reasoning and fine-grained spatial grounding. We introduce EyeVQA, a unified visual question answering
benchmark for comprehensive evaluation of ophthalmic VLMs. EyeVQA is constructed from 21 available ophthalmic
datasets and contains 20,000 question-answer pairs spanning six disease groups and seven question types: Single-Choice, Multi-Select, Variable-Select, True-False, Ranking, Point Location, and Bounding Box. Gold answers are deterministically derived from source-provided diagnoses, severity grades, clinical findings,
segmentation masks, bounding boxes, and anatomical landmarks, enabling reproducible evaluation without relying on model-generated annotations. Notably, 44.5% of the questions require
reasoning across multiple images, extending evaluation beyond conventional single-image medical VQA. We
benchmark fourteen representative general-purpose, scientific, and medically specialized VLMs under a unified zero-shot protocol. The best-performing model only achieves an overall score of 62.8, while substantial gaps remain in spatial grounding and cross-task generalization. These results highlight the limitations of current VLMs in comprehensive ophthalmic visual understanding and establish EyeVQA as a diagnostic
benchmark for developing more reliable and spatially grounded ophthalmic
multimodal models. The project page is available at https://github.com/PKUTHM/EyeVQA.
Read the full abstract ▾
All authors Gujie Shao, Zixun Xie, Xuechun Xing, Ruixiang Wang, Ziyun Lan, Yanlin Qi, Gangyi Zhang, Yuxin Yang, Dawei Li, Haiming Tang
Artificial Intelligence
From Trajectories to Grounded Preferences: Process Preference Synthesis via Interaction Element Graphs for Web PRMs
Yangzhe Peng, Xiaoyang Wang, Yiyang Zhao et al. (+2 more)
Comparative Process Reward Models (PRMs) provide critical step-level guidance for autonomous web
agents by evaluating state-conditioned preferences between candidate actions. However, existing preference training data synthesized via multi-policy sampling suffers from a severe scarcity of Grounded Minimal Contrastive Pairs (GMCPs)-where competing candidates target genuine on-page elements with identical action types. In representative baselines preference data (namely, WebArbiter), GMCPs account for merely 24.19%, biasing PRMs during training to rely on shallow shortcuts (such as element
hallucinations and action type mismatches) rather than acquiring genuine contextual decision semantics. To address these challenges, we propose SURFPRM, a graph-guided process preference synthesis framework for comparative Web PRMs. SURFPRM structures web demonstrations into a persistent Interaction Element Graph that acts as an environment-grounded negative action proposal mechanism, systematically synthesizing contrastive negative actions across spatial, temporal, and spatiotemporal confusion axes. This elevates the GMCP proportion from 24.19% to 74.60%, producing the curated SURFPRM-DATA
dataset. Across six open-source backbones (3B to 9B parameters), PRMs trained on SURFPRM-DATA outperform baseline-trained models on average on WEBPRMBENCH and rival leading proprietary LLMs. In downstream reward-guided trajectory search on WEBARENA-LITE, SURFPRM provides step-level guidance for both GPT-4o (+14.21%) and GPT-4o-mini (+12.83%) policies, yielding substantial improvements in complex web task success rates.
Read the full abstract ▾
All authors Yangzhe Peng, Xiaoyang Wang, Yiyang Zhao, Lijun Wu, Kun He
Artificial Intelligence · Cross-listed from cs.LG
Analog-Friendly Predictive Coding without Activation Derivatives
Francesco Innocenti
Predictive coding (PC) is a local, energy-based alternative to backpropagation (BP) whose iterative inference dynamics make it attractive for implementation on analog hardware. However, standard nonlinear PC requires evaluating the derivative of the activation function during both inference and learning, which can be difficult to realise physically. Here, we introduce activation-matched Bregman PC, replacing standard squared-error energies with Bregman divergences matched to the activation function. This formulation eliminates activation derivatives and, when combined with inference via mirror descent, yields local inference and learning rules requiring only weighted sums, local prediction errors, state integration, and the activation function. In digital experiments, Bregman PC performs comparably to standard PC and BP on classification and generative tasks, while preserving characteristic learning dynamics of PC and its convergence to BP under stable large-model parameterisations. These results provide a more analog-friendly formulation of nonlinear PC while retaining its key computational properties.
Read the full abstract ▾
Artificial Intelligence
ALLOT: Budgeted Hybrid-Memory Routing for Knowledge Updates in LLMs
Shanfeng Huang, Zhou Fang, Song Xiao et al. (+1 more)
For large
language models (LLMs), parametric adaptation is costly when
retrieval already suffices. We introduce ALLOT, a hybrid-
memory routing framework that separates learned write priority from a hard parametric budget. A
memory-aware router combines frozen text representations,
retrieval confidence, and relation metadata; a single ranking supports multiple write budgets while preserving all facts in external
memory. On CounterFact with Qwen3-4B, ALLOT reaches 0.760 accuracy at a 20% parametric-write budget and recovers 78.4% of the budget-matched oracle gain, with 80% fewer parametric writes than dual-writing every fact. At this budget, jointly adding
retrieval and relation features to text improves normalized oracle gain by 6.2 percentage points. Complementary Qwen3-0.6B shared-store results achieve dual-write-level accuracy with 6-14.5% parametric writes, and cross-
benchmark transfer retains approximately 88% of in-domain gain. These results support allocating adaptation capacity according to its incremental value rather than treating every factual update as an equally valuable training target.
Read the full abstract ▾
All authors Shanfeng Huang, Zhou Fang, Song Xiao, Hai Du
Artificial Intelligence
Enabling Timely Guidance before Skill Retrieval: Retaining Helpful Warm Tips in Agent Context
Feng Liang, Yupeng Li, Runhao Zeng et al. (+2 more)
Reusable skills help LLM-based
agents solve complex tasks, but the
agent must receive guidance before it commits to an ineffective approach. Existing skill mechanisms often expose only metadata and load full content on demand, leaving useful guidance unavailable until the
agent decides to retrieve it. General
memory methods can incur substantial maintenance overhead, while keeping guidance in conversation context risks repeatedly exposing the
agent to irrelevant or harmful advice. We propose TipsWarm, a mechanism that complements existing skill mechanisms by maintaining a budgeted pool of skill-derived keypoints, or warm tips, for selective injection into the context of every message turn. By separating event-triggered LLM assessment from inexpensive per-turn screening, it makes transferable skill guidance readily available while controlling maintenance costs. In three coding and iterative task-execution
benchmarks, TipsWarm achieves the highest task success rate while remaining time-efficient, compared to recent skill and general
memory baselines.
Read the full abstract ▾
All authors Feng Liang, Yupeng Li, Runhao Zeng, Francis C. M. Lau, Xiping Hu
Artificial Intelligence · Cross-listed from cs.CV
Progressive-View On-Policy Distillation for Regional-to-Global Transfer in Multimodal LLMs
Shanfeng Huang, Zhou Fang, Song Xiao et al. (+1 more)
Regional-to-global
distillation uses crop-conditioned guidance to improve full-image understanding. The challenge is to effectively transfer the teacher's crop-based advantage to the student's full-image inference. We propose progressive-view on-policy
distillation (PVD), which shifts the student's view distribution from the crop toward the full image through an intermediate aspect-preserving padded crop. The padded crop preserves regional content while matching the full image's visual-
token grid. Across stages, the view mixture assigns increasing probability to the full image. A lightweight regional-advantage weighting reallocates
token-level supervision using the crop-conditioned teacher-student log-probability gap. Evaluated under each sampled input, it applies mild reweighting when the gap is small and emphasizes higher-gap
tokens when the gap widens. A Jensen-Shannon metric decomposition interprets this schedule as a shift from matched-input imitation toward the deployment objective. Across
benchmarks spanning perception, visual mathematics and general
multimodal question answering, PVD-full reaches an average accuracy of 77.51 over three seeds, improving on the reward-free
distillation baseline by 2.01 points and on its reward-matched variant by 1.00 point. In the reward-free setting, PVD-distill still gains 1.16 points.
Read the full abstract ▾
All authors Shanfeng Huang, Zhou Fang, Song Xiao, Hai Du
Artificial Intelligence
HyperReCo: Retrieving and Connecting Evidence with Hypergraph Neural Networks for LLM Multi-hop Reasoning
Zicheng Zhao, Linhao Luo, Junnan Dong et al. (+4 more)
Large
language models (LLMs) have shown strong capabilities, with
retrieval-augmented generation (RAG) supporting complex multi-hop
reasoning by retrieving evidence distributed across documents. Graph-based approaches exploit connections among evidence, and hypergraph-based
retrieval further preserves higher-order entity associations within documents and connects documents through shared entities. However, existing hypergraph retrievers often rely on predefined structural expansion or
diffusion, which may miss query-dependent interactions needed to identify relevant evidence. They also leave connections among retrieved evidence implicit, requiring LLMs to reconstruct these connections before
reasoning. Therefore, we propose HyperReCo, a framework for retrieving and connecting evidence with a hypergraph neural network (HyperGNN). We represent each document as a hyperedge over its extracted entities, with shared entities connecting the hyperedges. Through hypergraph message passing with joint supervision over documents and entities, the HyperGNN learns query-dependent interactions to retrieve complementary evidence. We further introduce Gradient-Guided Hyper-Path Decoding (GGHD), which uses gradient attribution to interpret the learned interactions and translate them into explicit hyper-paths that help LLMs combine complementary facts for multi-hop
reasoning. Experiments on six
benchmarks show that HyperReCo achieves the best
retrieval performance among the compared methods on all three multi-hop QA
datasets, together with strong downstream QA performance. Case studies and further analyses demonstrate the utility of decoded hyper-paths for connecting retrieved evidence.
Read the full abstract ▾
All authors Zicheng Zhao, Linhao Luo, Junnan Dong, Haoran Luo, Xiaoli Li, Shirui Pan, Chen Gong
Artificial Intelligence
RLHarness: Co-evolving Procedural Skills with Reinforcement Learning for Long-horizon Multimodal Reasoning
Ziqiao Shang, Zian Xu, Ji-Chen Yan et al. (+6 more)
Multimodal reasoning requires models to preserve visual evidence through long decision chains while selecting appropriate procedures across diverse scenarios and rules. When learning is guided only by terminal verifiers, reinforcement learning (RL) reveals whether a final answer is correct but not how it should be produced. The policy must therefore discover reusable
reasoning procedures while learning to execute them, creating a program cold-start problem. Skills can externalize successful procedures, reduce repeated exploration, and provide inspectable guidance. However, a fixed Skill Bank assumes that this guidance remains compatible with an evolving policy, while updating Skills alone can leave their triggers, execution protocols, and demonstrations stale or mutually inconsistent. We introduce RLHARNESS, which organizes Skills, selection and execution protocols, few-shot demonstrations, and task contracts into a unified, versioned Harness and alternates Harness
evolution with policy learning. An Exploration-
Distillation Harness builds the initial Harness and version-aligned verified traces for SFT and DAPO I. After the first RL block, a Post-RL Reconstruction Harness rebuilds Skills, protocols, and demonstrations from fresh success-failure rollouts, and DAPO II adapts the policy to the reconstructed program. RLHARNESS improves Accuracy from 16.25%/27.50% to 62.00%/50.00% on MetroMap/TravelMap and raises F1 score from 37.13%/45.50% to 65.81%/65.51% on Fee-VL/Cancel-VL. All four tasks achieve their best results only after reconstruction and DAPO II, showing that an evolving Harness complements RL by continually updating the external program that the policy learns to execute.
Read the full abstract ▾
All authors Ziqiao Shang, Zian Xu, Ji-Chen Yan, Weiming Wu, Ziyi Jia, Jie Meng, Tao Huang, Shan Huang, Lan-Zhe Guo
Artificial Intelligence · Cross-listed from cs.LG
Active Feature Acquisition With Incomplete Training Data
Reza Rezvan, Valter Schütz, Han Wu et al. (+2 more)
In many prediction tasks, acquiring all features can be a prohibitively expensive or outright impossible task. Further, in many cases a static subset of features may not be enough to solve the problem sufficiently across various instances. Active Feature Acquisition (AFA) addresses these problems by formalizing the trade-off between feature cost and predictive performance during sequential feature selection. However, prior AFA work largely assumes access to complete training data, an assumption that is often violated in practice. Here we study AFA with Incomplete Training Data (AFA-ITD), showing that under missing completely at random (MCAR) data, one-step acquisition values remain unchanged, whereas multi-step values can decrease. We analyze three approaches to learning from incomplete data: aliasing, filtering, and generative restoration. We show that filtering can require a number of training instances scaling exponentially with the dimension, whereas generative restoration scales exponentially with the acquisition budget. We empirically test our theory on a controlled experiment and across common AFA
datasets and find that missingness mainly damages methods that exploit multi-step acquisitions and that generative restoration is able to recover lost performance in many experiments. Code is available at https://github.com/Linusaronsson/AFA-
Benchmark/tree/missing-data.
Read the full abstract ▾
All authors Reza Rezvan, Valter Schütz, Han Wu, Linus Aronsson, Morteza Haghir Chehreghani
Artificial Intelligence · Cross-listed from cs.LG
Superposed Inference for Hyperdimensional Computing
Quanling Zhao, Nilesh Prasad Pandey, Ye Tian et al. (+1 more)
Hyperdimensional computing (HDC) is attractive for efficient and robust learning, but conventional inference still encodes every query independently, repeatedly paying the cost of high-dimensional projection. We introduce SupHDC, a new inference paradigm that processes multiple queries through a shared encoding computation. SupHDC assigns lightweight random slot keys, superposes the keyed queries before encoding, and uses slot-specific classifiers to recover their individual predictions. A random-feature kernel view explains why exact recovery of each hypervector is unnecessary: inference only needs to preserve the class evidence that determines the prediction. Across ten
datasets, SupHDC achieves 1.39x analytical speedup with no average accuracy loss, and up to 2.08x speedup with only a 2.67 percentage-point mean accuracy loss. On a Raspberry Pi~5, it delivers 2.01x measured wall-clock speedup with a 2.26 percentage-point loss in mean prediction accuracy. SupHDC shows that high-dimensional redundancy can be used not only for robustness, but also as capacity for shared inference.
Read the full abstract ▾
All authors Quanling Zhao, Nilesh Prasad Pandey, Ye Tian, Tajana Rosing
Artificial Intelligence · Cross-listed from cs.LG
Phase Space Attention:A Hairer Lift Circumvents the Single-Layer Induction Obstruction
Kingsuk Maitra, Shagun Sood Morteza Hosseini, Suman Gunnala et al. (+1 more)
We circumvent the Sanford-Hsu-Telgarsky (SHT) single-layer induction obstruction within a linear, one-step, causal, bilinear, symplectically consistent design class on the post-RoPE substrate, by lifting
attention onto a
symplectic phase space, mirroring Hairer's lift of Stormer-Verlet. The lift exits the premise of the SHT counting argument rather than the bound itself. The reframing exhibits the obstruction as a filter-order gap: a one-layer bilinear score realises a $z$-transform of joint order $(0,0)$, whereas the induction discriminator requires key-side order $\geq 1$. Applying the
symplectic upper shear $M_\gamma:(q,p)\mapsto(q+\gamma p,p)$ to the post-RoPE query and key streams closes it. We prove this lift is unique within the factorised subclass, exactly
symplectic at operator level, and requires post-RoPE placement; and in an explicit $T_4$-only Gaussian reduction we derive a closed-form two-branch induction phase transition, held out at $r=0.9876$ with zero fitted parameters. That law is an analytically solvable limit, not a robust prediction: restoring the $T_3$ channel moves $\gamma_c$ at $d_k=64$ from 1.030 to 0.569 and removes the crossover. Deployability follows by exact derivation: the KV cache is unchanged, prefix reuse and speculative decoding are preserved, overhead is $6d$ FLOPs per
token per layer, INT8 headroom grows by at most $\log_2(1+2\gamma)$ bits, fused kernels are unmodified, and no parameters are added. At 91.3M parameters a supercritical sweep locates an emergence band: induction forms 3/3 seeds at $\gamma=0.80$ in a mean of 717 steps, against 2/3 seeds and 2700 steps at $\gamma=0$. Adverse results are reported as directly: a key-only half-lift reaches 0.949 against 0.811 for the symmetric operator, so if induction accuracy is the objective, the half-lift is the better construction. Forty-one notebooks and result files ship as ancillary material.
Read the full abstract ▾
All authors Kingsuk Maitra, Shagun Sood Morteza Hosseini, Suman Gunnala, Vikram Gupta
Artificial Intelligence · Cross-listed from cs.LG
What Can a Leaderboard Certify? Compositional Controllability for Fair Evaluation and Training of Biomedical Literature-Review Agents
Zhaowei Han, Xiang Zhang, Lingxiao Guan et al. (+4 more)
Leaderboards rank long-horizon
agents by their final outputs. Yet a higher score alone does not establish whether two systems are comparable or which stage accounts for the difference. Unequal evidence, inputs, or budgets can affect scores, and statistical corrections do not remove this mismatch. We introduce compositional controllability to address these questions. A comparison window covers one stage, several stages, or the whole
agent. Our central result bounds the gap between observed and controlled score differences using only nuisance outside the window. This yields an admissibility test applied before scores are inspected. Inadmissible comparisons are refused. For admissible pairs, an ordering is certified only when the score gap exceeds the combined sampling and nuisance radii; otherwise, it remains undecided. These decisions give each system a rank interval. We introduce BioLitBench, a
benchmark of 2,042 biomedical articles represented as structured claim graphs. Among seven published pipelines, a conventional statistical analysis declares a winner in 14 of 21 pairwise comparisons. Yet the top-ranked system alone received the target review's bibliography. To isolate pipeline performance, our test requires matched inputs and a fixed backbone model. It refuses 11 of the 21 comparisons, including every comparison involving the top-ranked system. Seven of the 14 conventional conclusions fall within these refused pairs. The same comparison windows support stage-level training. We train SCRIBE on Qwen3.8-27B using rewards measured at each stage's exit. Under matched evidence, SCRIBE achieves a certified rank interval of [1,2], with certified advantages over all evaluated published pipelines and the evaluated Claude and OpenAI
agents. Under same pool, SCRIBE matches the strongest published retriever and is certified above three published pipelines.
Read the full abstract ▾
All authors Zhaowei Han, Xiang Zhang, Lingxiao Guan, Danqi Hu, Kai Liu, Kevin Chang, Jie Liu
Artificial Intelligence
Delayed Supervision for Test-Time Language Models
Jinha Kim and Taksh Kothari
Test-time
language models adapt a compact
memory while processing the input sequence. This perspective encompasses nonlinear fast-weight learning in LaCT, associative delta-rule updates in DeltaNet, and generalized delta-rule state updates in RWKV-7. Training these models to predict the next
token does not explicitly require a fact to remain accessible after many subsequent
memory updates. We study delayed supervision for this test-time
memory: during post-training, ask a simulator-grounded question only after a long interval of unrelated events, and supervise its answer alongside ordinary next-
token prediction. Questions are evaluated on disposable branches, so their answers never enter the continuing event stream. The construction distinguishes retention from revision: a retained fact must remain valid throughout the delay, whereas a revised fact must be answered with its latest value. We evaluate this approach on LaCT-760M and plain DeltaNet-1.3B using TextWorld training trajectories and shared BABILong and RULER evaluation panels, and include a separately reported RWKV-7 comparison. Relative to event-only training, delayed QA improves BABILong by 5.48 percentage points for LaCT and 1.32 points for DeltaNet, and single-needle RULER by 1.45 and 3.27 points, respectively. The RWKV-7 comparison reports gains of 4.60 and 7.00 points on its own panels. These results support delayed semantic supervision as a practical outer training objective for usable test-time
memory, while leaving open how much of the benefit derives specifically from delay rather than general question-answering and answer-termination supervision.
Read the full abstract ▾
Artificial Intelligence
PhiFold: Towards Dynamic Protein Design with Physics-Structured Covariance Modeling
Yutian Liu, Mujie Lin, LanqianZhang et al. (+4 more)
Protein design is moving beyond structural correctness toward function-aware design, yet existing generative models typically treat dynamics as a downstream property estimated through simulation or prediction after structure generation. Using MD trajectories as a generative target is also undesirable because
stochastic, path-dependent trajectories over-specify the underlying
equilibrium ensemble. We introduce PhiFold, a framework for jointly generating
protein backbones and their second-order dynamics, represented by residue-displacement covariance. Rather than predicting the quadratically sized full covariance, PhiFold decomposes dynamics into three interpretable components: local flexibility, a low-rank collective-motion representation, and residue-wise collective participation. These components are assembled into a positive-definite covariance
matrix with exact marginal consistency, yielding a compact and physically constrained representation of
equilibrium dynamics. Across generated
proteins, PhiFold improves recovery of local fluctuations and long-range residue coupling while remaining competitive on dominant collective-motion subspaces. It further enables bidirectional control of residue flexibility while preserving backbone designability. By unifying structure generation with an explicit representation of
equilibrium dynamics, PhiFold lays a foundation for designing
proteins not only by how they look, but also by how they move.
Read the full abstract ▾
All authors Yutian Liu, Mujie Lin, LanqianZhang, Meng Fan, Chang Liu, ZhiweiNie, Siwei Ma
Artificial Intelligence
Train4Merge: A Controlled Single-Teacher Study of RL vs. SFT Teachers for OPD-Based Model Merging
Jingyuan Huang, Zuming Huang, Yucheng Shi et al. (+4 more)
Domain experts trained from a shared checkpoint can transfer their specialized capabilities to a single student through on-policy
distillation (OPD). Existing research primarily focuses on improving this merging process, while the
algorithms used to train the experts have received limited systematic comparison. We investigate which training
algorithm produces teachers better suited to OPD through controlled single-teacher comparisons of supervised fine-tuning (SFT) and reinforcement learning (RL) across Agentic,
Reasoning, and Perception. Teachers and students share the same Qwen3.5-9B initialization, and the two teacher types are compared at similar task performance. Our experiments show that RL teachers yield stronger students and higher recovery of teacher performance gains across all three domains. At their best checkpoints, RL-guided students outperform SFT-guided students by 4.27, 1.50, and 0.86 percentage points, respectively. In Agentic, the best SFT-guided student recovers only 44.44% of its teacher's performance gain over the base model, whereas the best RL-guided student recovers 115.00%, surpassing its teacher. Our further analysis shows that RL teachers undergo smaller parameter displacements from the shared initialization than SFT teachers. These findings support the hypothesis that RL teachers' smaller departures from the student's starting point facilitate learning through OPD, resulting in stronger students.
Read the full abstract ▾
All authors Jingyuan Huang, Zuming Huang, Yucheng Shi, Zhongzhi Li, Xiaoming Zhai, Wei Chu, Ninghao Liu
Artificial Intelligence
Agentsensus: Consensus-Compressed Shared Memory for Multi-Agent Story Worlds
Yu Pan
A agentic story world is a dynamic system simulating who learned what, when, and from whom – yet the standard design gives each character a private
memory stream. A shared event is therefore stored once per witness, large duplication will be incurred in terms of storage. We present Agentsensus, a story-world simulation framework in which there is an unified long-term
memory. Records of the same event merge into one owned by all its witnesses, and semantically relevant
memory records are linked. We evaluate on four worlds – two classical Chinese novels, Hamlet, and a real-world conflict timeline – run for 40 to 80 rounds against three per-character
memory designs under an equal-granularity protocol. Agentsensus writes 22-44% fewer entries than the closest baseline and is the only design whose
memory becomes shared (14-28% of records held by more than one character, some by 10) and linked (94-99%), at judged simulation quality indistinguishable or even better than the baselines. An ablation attributes this to the merge itself: disabling it multiplies the store by 3.1x and takes sharing to exactly zero. Sharing also compounds with the horizon rather than saturating early, rising 6% to 9% to 14% as one world is re-run at 10, 20 and 40 rounds.
Read the full abstract ▾
Artificial Intelligence
GLIDE: Generalized Layer-wise Intrinsic Distributional Evaluation for Heterogeneous LLM Agents
Wei Zhu, Yiming Wang, Rui Wang et al. (+3 more)
LLM
agents require reliable step-level evaluation to compare candidate branches and allocate computation effectively. However, lightweight evaluation remains challenging. External verifiers introduce additional inference cost, while
agent-produced confidence or self-evaluation scores can be miscalibrated, especially when candidates are generated by heterogeneous
agents. We propose Generalized Layer-wise Intrinsic Distributional Evaluation (GLIDE) for LLM
agents. GLIDE derives intrinsic step evidence from layer-wise residual coherence, which measures whether local residual updates consistently support the global residual change induced by a candidate step. It calibrates this evidence against the recent score distribution of the generating
agent and converts it into a pessimistic reward that jointly accounts for absolute residual evidence and
agent-relative standing. The reward provides a cross-
agent value signal for MCTS branch selection, while normalized predictive uncertainty guides adaptive branching. Experiments on multi-hop
reasoning, sequential decision making, and symbolic logic show that GLIDE improves task performance, step-level ranking quality, and computational efficiency without external verifiers or task-specific supervision.
Read the full abstract ▾
All authors Wei Zhu, Yiming Wang, Rui Wang, Lixing Yu, Kun Yue, Zhiwen Tang
Artificial Intelligence · Cross-listed from cs.CL
TRAP: Understanding and Mitigating Privacy Memorization in Language Models
Muhammed Ustaomeroglu, Ziyue Xu, Hanshen Xiao et al. (+3 more)
Fine-tuning a
language model on sensitive records can leave it able to reproduce them. We ask when this memorization arises and how to prevent it without knowing in advance which spans are sensitive. Our starting point is that most memorization scores and attacks share one statistical core: whether the model assigns a
token more probability than some reference would. Taking as the reference a model trained on the complementary half of the same corpus gives the Target Reference Advantage (TRA), a per-
token signal that separates what a model fit to a particular record from what it learned across records, and is cheap and differentiable. We then study what drives memorization during fine-tuning: it keeps growing well past the validation minimum, is larger on small
datasets and at higher learning rates, and higher when the underlying task is harder. Early stopping removes much of it, but because it is chosen by aggregate validation loss it helps least for rare, hard-to-predict spans embedded in otherwise learnable text, which is exactly what sensitive information tends to be. We therefore introduce TRAP, a one-sided penalty on tokenwise TRA that acts only where the target model pulls ahead of its reference. On student essays with annotated personal information and clinical cases with patient identifiers, TRAP brings memorization near the level of an untrained model at little utility cost, where generic regularizers barely move and differential
privacy gives up most of what fine-tuning bought.
Read the full abstract ▾
All authors Muhammed Ustaomeroglu, Ziyue Xu, Hanshen Xiao, Peter Cnudde, Guannan Qu, Holger R. Roth