跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+1084新增论文
+883其中 AI 相关
+130已有论文修订
+94新增 AI 导读

arXiv 最近完成扫描:2026-11-18 04:28 · 全站最近入库:2026-11-18 19:45各信源进度 →

统计口径与数据新鲜度

统计 2026-11-18 06:26 — 2026-11-19 06:26(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-18 04:28最近发现新增版本:2026-11-18 19:45库内论文最新更新时间:2026-11-18 08:00

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

26,870 已收录论文22,374 AI 相关8,156 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 22,374 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-18 04:27(北京时间)
新论文 · v1arXiv:2611.22304

Cross-Modal StateKV Injection for Communication in Multi-LLM Systems across Temporal Segments

Xiaoqian Zhang, Zhihao Zhang

原文摘要

Video understanding with multiple multimodal language models requires a communication interface that can transfer complementary visual observations. Text-based communication is interpretable, but introduces an intermediate generation stage and can omit information relevant to subsequent reasoning. We study Cross-Modal StateKV Injection, a provider--carrier framework that communicates through key--value (KV) states at visual-token positions. Provider models observe complementary video frames; their visual KV states are aligned and projected into the carrier representation space and integrated through gated residual updates. Injection preserves the carrier KV-cache sequence length, while retaining the additional computation required by the providers and the communication module. The reported experiments on MLVU and Video-MME compare carrier-only inference, text-based communication, and StateKV with two provider architectures. Results show modest, task-dependent accuracy gains and lower reported latency than the evaluated text-communication baseline, with some categories remaining unchanged or decreasing. Increasing provider-side frame coverage offers additional gains on selected tasks without extending the carrier cache. These findings motivate further study of visual KV communication under explicitly controlled observation and computation budgets.

cs.CV

新论文 · v1arXiv:2611.22302

Where Do Human-to-Robot Transfer Gains Come From?

Jiachen Li, Shihao Li, Soovadeep Bakshi, Longteng Fan, Dongmei Chen, Long Chen, Zhuo Xu

原文摘要

Human data can support robot policy pretraining, but gains attributed to a data recipe may also reflect subset size, training variability, or downstream adaptation. In an HPT implementation on a 20-task MetaWorld suite with three training blocks, a constrained random 5,000-recording subset matches most of the task-informed subset's gain over the 40,000-recording pool at 20,000 updates (3.69 of 4.29 percentage points), leaving equal-size selection differences of 0.60 points at 5,000 recordings and 0.45 points at 1,250 recordings, with wide paired intervals. Across 12 comparisons, task orderings repeat across seeds for several contrasts, but a fixed predictor built from the other two seeds trails the mean or difficulty reference in at least one held-out fold for every comparison. Neither mechanism probe shows its predicted positive interaction: task-slot mismatch yields an interaction of $-0.0347$ points after adaptation, and in a six-task observation experiment the observation-by-pretraining interaction changes sign between seeds ($-18.29$ and $+1.04$ points) while full-state pretraining gains are positive in both ($+6.37$ and $+2.31$ points). This retrospective study shows how equal-size and capacity controls, seed-level reporting, and targeted interventions change the interpretation of apparent transfer gains.

cs.RO

版本更新 · v2arXiv:2607.00444

Spatiotemporal and Multi-Robot Motion Planning on Graphs of Space-Time Convex Sets

Jingtao Tang, Zining Mao, Lufan Yang, Hang Ma

原文摘要

Decoupled multi-robot motion planning avoids the high-dimensional joint configuration space but requires repeated single-robot planning among moving obstacles, where route choice and timing are coupled. We develop a framework based on graphs of space-time convex sets (ST-GCSs) for continuous spatiotemporal planning and multi-robot coordination. An ST-GCS represents collision-free space-time by convex sets, allowing the sequence of regions and the geometry and timing of piecewise-linear trajectories to be optimized jointly under velocity limits. To incorporate moving obstacles and previously planned robots, we develop Exact Convex Decomposition (ECD), which removes their swept occupancies while maintaining a convex decomposition for subsequent planning queries. Building on path-based GCS search, we develop a best-first solver tailored to the ST-GCS motion model, exploiting temporal reachability for dominance pruning and using admissible heuristics with separate or joint search-key evaluation. Precomputed heuristic bounds can also be reused across reservation-updated graphs. We integrate ST-GCS planning and ECD into prioritized multi-robot planning and introduce windowed coordination with dynamic window adjustment to limit reservation-induced graph growth. Experiments demonstrate improved scalability and competitive solution quality against optimization-based and sampling-based planners on challenging spatiotemporal and multi-robot problems. Large-scale simulations coordinate up to $100$ robots within a minute, while real-robot demonstrations showcase the practical applicability of the framework. Project homepage: https://sites.google.com/view/stgcs.

cs.RO · cs.AI

版本更新 · v3arXiv:2610.29362

Compositional neural networks for multimodal X-ray and neutron scattering inversion

Valentin Munteanu, Alexander Hinderhofer, Dmitry Lapkin, Vladimir Starostin, Frank Schreiber

原文摘要

X-ray and neutron scattering are indispensable for structural characterization across condensed-matter physics and materials science. Yet, because scattering experiments generally record intensities without phase information, distinct structures can yield similar scattering patterns, rendering the corresponding inverse problems non-unique. This ambiguity can be reduced by combining complementary measurements, but their integration into machine-learning frameworks remains challenging. Here we introduce a modular framework for probabilistic multi-measurement analysis based on prior-amortized flow matching. Single-observation posterior estimators are trained separately for individual observation types or physical probes and combined only at inference time through a compositional sampler. This late fusion approach allows pretrained single-observation models to be reused without dedicated joint retraining. As a case in point, we consider reflectometry, which provides depth-resolved, nanometer-scale information about thin films and interfaces. Our single-observation demonstrations span X-ray, time-of-flight neutron and polarized neutron reflectometry. In separate co-refinement studies, task-matched estimators enable homogeneous composition across multiple neutron contrasts and heterogeneous multimodal composition of X-ray and neutron measurements. Together, these results establish compositional flow matching as a modular approach to uncertainty-aware inversion and co-refinement, with potential applicability to other scattering problems in which multiple observations constrain a common physical structure while retaining measurement-specific parameters.

cs.LG · cond-mat.mtrl-sci · physics.app-ph

新论文 · v1arXiv:2611.22301

MCTS Is Counting the Wrong Thing: Learning Exploration Counts for LLM Reasoning

Xuan Zhang, Zhijian Zhou, Zicheng Hu, Yan Feng, Hongyan Hao, Yuan Qi, Chao Qu

原文摘要

Monte Carlo tree search (MCTS) has powered landmark successes in reinforcement learning, yet its gains in LLM reasoning can remain modest despite substantial search effort. We identify an overlooked bottleneck: what MCTS counts. Standard UCT assigns independent visit counts to textual branches, allowing different formulations of the same reasoning strategy to repeatedly receive exploration bonuses. Our theoretical analysis shows that this count fragmentation can impair exploration efficiency, as quantified by increased cumulative regret. To address this problem, we introduce Learned Pseudo-Count MCTS (LPC-MCTS), which changes only the exploration statistic in UCT. A lightweight Coin-Flipping Network head on the policy's hidden representations estimates representation coverage, enabling related branches to share exploration evidence. Across 5 mathematical-reasoning and 7 search benchmarks, LPC-MCTS consistently outperforms strong MCTS baselines. Holding each LPC-MCTS-trained policy fixed, replacing standard MCTS counts at inference improves average mean@16 accuracy across 5 mathematical benchmarks by 3.79 percentage points on Qwen2.5-7B and 2.71 points on Qwen3-4B. Applying LPC-MCTS during both training and inference yields gains of 4.96 and 4.37 points over standard MCTS, respectively. Together, our theory and experiments show that changing what MCTS counts can substantially improve its exploration efficiency and reasoning performance.

cs.AI

版本更新 · v2arXiv:2611.21038

ScoutCache: Efficient Few-Step Autoregressive Video Diffusion via Conditional Readout and Selective Caching

Hao Tan, Jinghui Qin, Xuming Hu

原文摘要

Few-step autoregressive video diffusion enables efficient long video generation, but inference still incurs substantial computational costs from repeated denoising and KV cache operations. Existing acceleration methods reduce parts of this overhead, while historical KV reading and cache construction remain costly across video chunks. To address these challenges, we propose ScoutCache, a training-free framework that jointly optimizes denoising, historical KV reading, and cache construction. We observe that shallow feature changes provide useful information for estimating clean latent updates, while historical KV dependence varies across Transformer layers and attention heads but remains relatively stable across video chunks. Based on these observations, we develop Conditional Denoising Readout (CDR) to estimate intermediate denoising updates from shallow feature changes and known noise variations. We further introduce Output Risk Guided Historical KV Reading (ORHR) to allocate historical reading budgets based on attention output perturbations, and Content and Coverage Guided Sparse Cache Construction (CCSC) to reduce cache construction costs by selecting informative spatial tokens while maintaining spatial coverage. Extensive experiments on Self-Forcing and LongLive using VBench and MovieGen demonstrate that ScoutCache achieves up to 2.28$\times$ and 1.86$\times$ speedups, respectively, while consistently improving generation quality without modifying pretrained model parameters.

cs.CV

版本更新 · v3arXiv:2611.15011

Fast Inference-Time Policy Refinement via Action-Affine World Models

Run Wang, Yilin Mo, Liang Wu

原文摘要

Imitation policies can fail through compounding errors, while existing refinement methods often require additional on-policy learning or expensive online planning. Our key insight is to model the response to small corrections around a policy proposal in an affine form that enables fast optimization. We refine a frozen imitation policy using an \textit{action-affine world model} (A\textsuperscript{2}WM), trained offline from perturbed policy rollouts. Conditioned on history and the proposal, the model directly predicts a nominal trajectory and a linear action response over the full horizon. With quadratic costs and linear constraints, refinement becomes a cheap quadratic program (QP), requiring only one model evaluation and one QP solve per control step. To the best of our knowledge, we are the first to combine an action-affine world model with QP-based inference-time refinement of frozen imitation policies. Across four manipulation tasks, our framework achieves the highest success rate among evaluated baselines, adding only 4.2 ms of mean refinement latency per control step, excluding base-policy sampling.

cs.RO

新论文 · v1arXiv:2611.22299

Harnesses That Improve Themselves: A Unifying Framework for Self-Improving Agent Harnesses

Deep Shah, Nehal Kathrotia

原文摘要

Whether a foundation model succeeds as an autonomous agent depends as much on its runtime software as on its weights. This runtime, the agent harness, manages context and memory, steers multi-turn control, executes tool actions in sandboxes, and verifies work before stopping. Because models differ in how they use tools and how they fail, a static harness tuned for one model or task often stalls, wastes tokens, or regresses on another. This bottleneck has driven a shift from prompt tuning to self-improving agent harnesses: systems that inspect their own execution traces, separate runtime bugs from model limits, and rewrite their own code. We present a unifying architectural and design framework for self-improving agent harnesses. The framework pairs four editable runtime components (context and state, agent control, action interfaces, and verification and recovery) with a four-stage loop that diagnoses failures, proposes code patches, validates candidates, and deploys updates. Using this framework, we show how optimizers trace errors across long trajectories and discover reusable runtime mechanisms such as pre-stop verifiers, loop breakers, and persistent memory. We further show why full-harness rewrites overfit or break across modules, and how modular updates with validation gates generalize across models while bounding token cost. To keep self-modification safe, we establish practical defenses against recurring failure modes including prompt bloat, benchmark gaming, and evaluator tampering. Finally, experiments on SWE-bench Lite validate this framework, showing that gated modular evolution outperforms prompt reflection and full-harness rewrites across models and repositories at a fraction of the token cost.

cs.AI

新论文 · v1arXiv:2611.22298

Streaming On-Policy Distillation: Decoupling Three Clocks for Stage-Wise Supervision Consumption

Yun Zhu Da Wang

原文摘要

In on-policy distillation (OPD), response completion, supervision delivery, and parameter updates need not share the same granularity. We develop streaming OPD by decoupling these three clocks: a response repeatedly delivers newly generated segments while generation continues, retaining the full causal prefix at each delivery and computing loss only on new positions. A consumption protocol then organizes supervision budgets, position selection, staleness handling, and update scheduling, allowing control over quantity, coverage, and version lag across training stages. In GSM8K experiments distilling Qwen3-4B into Qwen3-0.6B with VERL, supervised-budget matching reduces the early gap between best test accuracies over seven common candidates from 4.55 to 2.42 points. Synchronous continuation from the same checkpoint reaches 73.39% best test accuracy over seven common candidates at u=384,...,576, versus 74.30% synchronous and 71.95% budget-matched (differences -0.91 and -2.35 points); an unpaired extended record reaches 76.12%. Freshness filtering and disagreement-overlap selection improve local peaks but do not establish sustained advantages. A system prototype shows lower data-supply waiting times without establishing an equal-update speedup. Quality recovery and supply-side benefits are assessed separately; end-to-end gains must account for continuation cost.

cs.AI

新论文 · v1arXiv:2611.22297

FIRE-Drive: Feasibility-Informed Risk Evaluation for End-to-End Diffusion Planning

Qifeng Xie, Kai-Yuan Feng, Wenkang Wan, Yue Wu, Maoguo Gong, Kai Sheng

原文摘要

Diffusion-based end-to-end autonomous driving models have shown strong potential in generating multimodal planning trajectories. However, their performance still depends heavily on trajectory initialization, mode evaluation, and supervision quality. Weakly constrained noisy initialization may introduce infeasible proposals and increase the burden of denoising. Moreover, selecting the optimal mode remains challenging when candidate trajectories are geometrically similar but differ in safety, interaction risk, and progress. To address these issues, we propose FIRE-Drive, a feasibility-informed risk evaluation framework for end-to-end autonomous driving. FIRE-Drive improves diffusion-based planning through dynamic feasible anchor initialization, candidate-specific bird's-eye-view risk evidence, and compact high-quality candidate supervision with lightweight refinement. This design aligns the initial support, risk-aware scoring, and multimodal supervision of trajectory modes under a unified planning objective. It enables the diffusion decoder to focus on refining plausible trajectories while the scorer distinguishes modes using local risk evidence. Experiments on NAVSIM v1 show that FIRE-Drive achieves a Predictive Driver Model Score of 91.5 with a ResNet34 backbone. Additional closed-loop evaluation on 220 Bench2Drive routes yields a Driving Score of 71.43 and a Success Rate of 42.73%. Component ablations and a scorer comparison under matched candidate inputs support the contributions of feasible initialization, auxiliary supervision, and risk-informed selection. Our code is available at https://github.com/Latte00/FIRE-drive.

cs.RO · cs.AI

新论文 · v1arXiv:2611.22296

SpecSearch: Towards Hardware-Aware Efficient Speculative Decoding

Selin Yildirim, Ali Zeyrek, Deming Chen

原文摘要

Speculative decoding (SD) accelerates large language model (LLM) inference, but its optimal configuration depends jointly on workload characteristics, GPU hardware, and available memory. For example, on the RTX~5090, the state-of-the-art speculative drafter DFlash achieves $3.50\times$ the throughput of autoregressive (AR) decoding at low concurrency, but only $0.76\times$ at high concurrency with short outputs. We present SpecSearch, a hardware-aware search engine that jointly optimizes decoding strategy, draft length, and precision across workload--GPU pairs. Its analytical cost model captures memory and compute constraints to identify feasible configurations and narrow the search space for throughput optimization. Guided by this model, calibration measurements are collected from multiple GPUs. Using the measurements, a Jev-based predictor learns to select high-throughput decoding configurations on held-out GPUs without requiring deployment-GPU calibration. The predictor is parameter-efficiently trained on calibration data, while adapting a frozen Qwen3-4B using LoRA adapters and a linear correction layer. Across NVIDIA RTX~5090, H100, L40 and AMD MI300X GPUs, SpecSearch with full calibration achieves an average of $2.01\times$ the throughput of AR and $1.36\times$ the throughput of SD, both served by vLLM. The proposed Jev-based predictor closely matches this search performance with $1.30\times$ throughput improvement over SD with vLLM, while eliminating the need for held-out GPU calibration. Additionally, it retains $97.36\%$ of the throughput achieved by SpecSearch that utilizes full calibration.

cs.AI · cs.DC

新论文 · v1arXiv:2611.22295

Similar Is Not Useful: Selecting In-Context Demonstrations by Closed-Loop Utility

Jiachen Li, Zhigen Zhao, Shihao Li, Dongmei Chen, Long Chen, Zhuo Xu

原文摘要

Retrieval-augmented vision-language-action (VLA) policies fill their context with the demonstrations whose frames look most like the current observation under frozen DINOv2 embeddings. We ask whether visual similarity selects the demonstrations that make the policy succeed. For a frozen retrieval-augmented VLA (RICL on $π_0$-FAST, LIBERO), we make the demonstration pool swappable at run time and score each demonstration's utility by regressing the closed-loop success of 60 fixed-size pools on their membership. On initial states disjoint from those used for scoring, pools of the five highest-utility demonstrations succeed on 66.4% of 760 paired episodes across 19 tasks, pools of the five most similar on 55.7%, and random pools on 55.5%. Utility beats similarity by 10.8 points ($p=3.2\times10^{-6}$; positive on 15 of 18 untied tasks), including on 12 tasks the policy was never primed on, and similarity-selected pools do no better than random ($+0.1$ points). Within a task, a pool's visual similarity to the test scenes tracks its success at Spearman 0.10; across tasks, similarity separates tasks whose scenes differ and falls to chance when two tasks share a scene. The gain comes from on-task closed-loop evidence: best-of-ten direct search over pools recovers two-thirds of it with 100 rollouts. Against the shipped 20-demonstration pool, utility-curated pools of five lead by 7.0 points over 560 paired episodes from two evaluation runs ($p=0.006$), and utility-ordered retrieval inside the full pool leads by 4.5 points ($p=0.055$).

cs.LG

新论文 · v1arXiv:2611.22294

Perception into Imagination: Executable Memory for World Models

Xianyao Li, Ruitong Tian, Rui Min, Fang Xu, Jing Du

原文摘要

Physical agents must use what they previously saw or touched after visibility or contact is lost. Their actions change the implications of these remembered facts. We introduce executable perceptual memory (EPM), a world-model framework that advances explicit perceptual state alongside neural latent dynamics during imagination. Its evolving read-out informs reward prediction, value estimation and policy learning. We align reward inputs with the transitions that produced them and supervise the reward head on open-loop rollouts of recorded actions. On Memory Maze $13\times13$, EPM reaches 30.37 return at 1M interactions, compared with 23.37 when the same evolving read-out informs only policy and value on a common perceptual front end: a gain of 7.00 [3.80, 10.21]. The complete method also improves control on $9\times9$ and in MiniWorld. Probes of remembered, out-of-view arrivals show stronger reward anticipation in both visual implementations. Concentration, tactile recovery and real tactile frames demonstrate complementary roles of remembered facts in imagined-policy training, motion-dependent decisions and sensory-content retention. EPM connects persistent perception to the predicted consequences and learning signals of a world model for Physical AI.

cs.LG · cs.AI · cs.CV · cs.RO

版本更新 · v4arXiv:2512.18857

CORE: Concept-Oriented Reinforcement for Bridging the Definition-Application Gap in Mathematical Reasoning

Zijun Gao, Zhikun Xu, Xiao Ye, Ben Zhou

原文摘要

Large language models (LLMs) often solve challenging math exercises yet fail to apply the concept right when the problem requires genuine understanding. Popular Reinforcement Learning with Verifiable Rewards (RLVR) pipelines reinforce final answers but provide little fine-grained conceptual signal, so models improve at pattern reuse rather than conceptual applications. We introduce CORE (Concept-Oriented REinforcement), an RL training framework that turns explicit concepts into a controllable supervision signal. Starting from a high-quality, low-contamination textbook resource that links verifiable exercises to concise concept descriptions, we run a sanity probe showing LLMs can restate definitions but fail concept-linked quizzes, quantifying the conceptual reasoning gap. CORE then (i) synthesizes concept-aligned quizzes, (ii) injects brief concept snippets during rollouts to elicit concept-primed trajectories, and (iii) reinforces conceptual reasoning via trajectory replacement after group failures, a lightweight forward-KL constraint that aligns unguided with concept-primed policies, or standard GRPO directly on concept-aligned quizzes. Across several models, CORE delivers consistent gains over vanilla and SFT baselines on both in-domain concept-exercise suites and diverse out-of-domain math benchmarks. CORE unifies direct training on concept-aligned quizzes and concept-injected rollouts under outcome regularization. It provides fine-grained conceptual supervision that bridges problem-solving competence and genuine conceptual reasoning, while remaining algorithm- and verifier-agnostic.

cs.AI · cs.LG

新论文 · v1arXiv:2611.22291

When the World Is the Test Suite: Agentic Intelligence for the Physical World

Kaixin Li, Meng Luo, Yanlin Li, Yuchen Tian, Xiang Liu, Xiangru Jian, Yan Yang, Zhiheng Xi 等 12 位作者

原文摘要

Coding agents built to repair software now write, test, and repair programs for robots, laboratories, and industrial controllers. Unlike in software engineering, no test suite comes with the code. Whether a program works must be established by running it on a robot, in a physics simulator, or through the policies it helps to train, and such evidence is costly to obtain and can itself be wrong. As these agents take on more physical tasks, how their programs are judged matters as much as how they are written. This paper examines this emerging area, agentic intelligence for the physical world, by treating the world as the program's test suite. We organize the discussion around three questions: how agents write programs for the physical world, how these programs are tested, and how agents learn from the results. We first trace how language models for robots and software engineering agents came together, ask which software testing practices carry over to physical tests, and describe each program by where it sits relative to the actuator and by what judges it. For writing, we cover programs that act as robot policies and programs that support policy learning by producing rewards, tasks, and training worlds. For testing, we examine what physical trials cost, what traces, learned judges, simulators, and formal checks can establish in their place, and what safety requires beyond task success. For learning, we discuss how tested programs are retained as experience, in skill libraries and in model weights. Finally, we turn to benchmarks and evaluation practice and outline open challenges. Writing programs is becoming cheap, while trustworthy evidence about what they do remains costly. Progress will therefore depend on a physical counterpart of continuous integration: verification cheap enough to run often, reliable enough to trust under search, and maintained alongside the experience it validates.

cs.RO · cs.AI · cs.SE

版本更新 · v4arXiv:2503.23326

Exploring Explainable Multi-agent MCTS-minimax Hybrids in Board Game Using Process Mining

Yiyu Qian, Tim Miller, Liyuan Zhao

原文摘要

Monte-Carlo Tree Search (MCTS) is a family of sampling-based search algorithms widely used for online planning in sequential decision-making domains and at the heart of many recent advances in artificial intelligence. Understanding the behavior of MCTS agents is difficult for developers and users due to the frequently large and complex search trees that result from the simulation of many possible futures, their evaluations, and their relationships. This paper presents our ongoing investigation into potential explanations for the decision-making and behavior of MCTS. A weakness of MCTS is that it constructs a highly selective tree and, as a result, can miss crucial moves and fall into tactical traps. Full-width minimax search constitutes the solution. We integrate shallow minimax search into the rollout phase of multi-agent MCTS and use process mining technique to explain agents' strategies in 3v3 checkers.

cs.AI

版本更新 · v2arXiv:2610.00445

One pool, many targets: a conservation layer and what archival data can identify

Zahra Khodagholi, Niloofar Yousefi

原文摘要

Pairwise guide-transcript scores do not enforce conservation of a finite guide-loaded RISC pool when they are interpreted independently as occupancies. We formulate a differentiable scalar equilibrium layer: one conservation equation with a unique positive root and exact implicit gradients. It yields a redistribution theorem, a qualified high-resource limit, an analysis of the retrieval approximation, and a conditional rank-invariance result: within one construct at one dose, rankings by fractional occupancy cannot distinguish equilibrium from independent scoring. We therefore audit the two experiments that proposition leaves open, dose and cross-context, on archival off-target data. Corrected thermodynamic affinities associate weakly with measured repression in the direction a working predictor requires, but a paired permutation test and a construct-cluster bootstrap do not establish added predictive value from the coupling: what survives their differing permutation-null baselines is 0.0026, which is 2.55 paired-null standard deviations from that baseline (p=0.010) and favours independent scoring. The dose fits are heterogeneous and frequently violate the model-implied exponent constraint, which is superlinear rather than sublinear, so these data do not identify the competition parameter. A saturable compression of the competitor set holds both accuracy targets on held-out guide families but is not faster at the size measured. The contribution is a reusable conservation operator and the experimental information needed to test it. Code, result files and the data manifest: https://github.com/shadi97kh/One-Pool-Many-Targets

cs.LG

新论文 · v1arXiv:2611.22288

ConfGuard: a physics-grounded reference monitor for predictive-maintenance large-language-model agents

Arunkumar V, Gangadharan G. R., S. Senthilkumar

原文摘要

Large-language-model agents now close the industrial predictive-maintenance loop, reading sensor windows, retrieving procedures and dispatching work orders. Two attack channels meet there: injected text in a retrieved document can hijack the decision, and a bounded sensor perturbation can corrupt the remaining-useful-life estimate behind it. Recent agent defenses police data provenance and tool privilege but cannot tell whether a prognostic quantity is physically plausible. ConfGuard is a reference monitor with a physics-grounded policy. Its conformal monitor tests each prediction, anchored to the asset's admitted history, against degradation physics: capped remaining life never rises and falls at most one cycle per cycle. Calibrating its threshold over whole asset trajectories lets the operator set the expected share of a clean asset's windows the running monitor escalates. A physics-policy validator checks the agent's maintenance-urgency decision against the decision a deterministic policy derives from the same interval. On all 10,196 C-MAPSS test windows the monitor cuts a covert attacker's reach on healthy assets from 0.529 to 0.046 and closes suppression of failing assets (0.151 to 0), escalating 0.027 of clean windows against a 0.05 target, while the fixed physical range it replaces leaves both attacks open. Behind a Qwen3-14B planner the composed monitors hold urgency-class injection to 0.00 and the universal sensor attack to 0.12, against 0.29 and 0.54 for a defended prompt. An adaptive attacker defeats a Mahalanobis detector on 0.504 of windows and adversarial training on 0.485. We report an attack class the projection misses.

cs.AI · cs.CR · cs.LG

新论文 · v1arXiv:2611.22287

Decoding Without a Decoder: Direct Pixel Prediction from Diffusion Transformers

Hansheng Chen, Shengqu Cai, Jan Ackermann, Rayhan Zirvi, Gordon Wetzstein, Xinchen Yan, Leonidas Guibas

原文摘要

Latent diffusion models rely on a decoder to map denoised latents to pixels, adding complexity and inference cost. We introduce Decoding Without a Decoder (DWD), which lets a finetuned diffusion transformer produce final pixels directly through a linear head. Unlike previous methods, where the denoising output leads to final pixels, DWD separates the two roles: the original denoising head continues to drive sampling, then the new DWD head predicts the pixels at the final step. We extend pretrained latent models with DWD efficiently by finetuning only the last few transformer blocks, training the DWD head with pixel quality losses while anchoring the denoising head to the pretrained model to preserve the sample trajectory. Across latent image and video generators, DWD removes the decoder and reduces inference cost while maintaining or even improving pixel quality. On Wan video generation, it visibly reduces detail distortion and flickering, lowering warping error by 23%. Remarkably, the same separation also improves pixel diffusion models that do not even use a decoder: a final-step DWD head produces substantially better texture detail in text-to-image generation, and improves ImageNet FID from 1.55 to 1.43, a new state of the art among pixel flow models.

cs.CV

新论文 · v1arXiv:2611.22286

Learning Regions of Attraction for Transverse Feedback Linearization under Controller Singularities

Abdullah Waqar Malik, Hamza Tariq, Adeel Akhtar

原文摘要

Transverse feedback linearization (TFL) achieves path following by exactly linearizing the error dynamics, but its controller is undefined on a singular set, such as zero speed for a unicycle or zero thrust for a quadrotor. The constrained region of attraction, the set of initial states from which the closed loop converges to the path without violating state or regularity constraints, determines where the controller can be safely activated. Exploiting the exact linear error dynamics, we show that this region is the largest forward invariant subset of the admissible set. We then decompose its state-constrained Zubov value function into an analytically known quadratic cost-to-go and a safety value function. We learn only the latter, by fitted value iteration without infinite-horizon trajectory labels. The resulting estimate combines an inner layer verified by interval branch and bound with a learned outer layer tested on samples. On a unicycle following a circle and a sinusoid and on a planar quadrotor, it covers on average 72.0%, 83.4% and 19.1% of the simulated region of attraction. In comparison, the quadratic cost-to-go and two physics-informed neural network (PINN) solvers of Zubov's equation cover at most 11.3%, 4.0% and 0.0% on average. Code and animations are available at https://gradslab.github.io/roa_tfl

cs.AI · eess.SY

版本更新 · v2arXiv:2610.23051

Agentic In-Context Learning: What Algorithms Can Transformers Learn in Closed-Loop Interaction?

Shiguang Wu, Yaqing Wang, Quanming Yao

原文摘要

The learning algorithm view of in-context learning (ICL) studies how fixed-parameter Transformers learn from supplied examples. Language model agents extend this setting through interaction: their actions determine what evidence enters context next. We call this regime Agentic In-Context Learning (Agentic ICL). Existing studies largely depend on externally supplied contexts or reward-bearing interaction histories, while work on general tool-using agents lacks a learning algorithm view. We represent an interactive learner by its state, update, and decision rule, covering active learning, bandits, and online reinforcement learning (RL). For expressivity, we construct causal Transformers that approximate these algorithms over a finite horizon, and bound the resulting loss in return. For training, we characterize when optimizing process rewards preserves behavior that maximizes the final outcome. Distillation experiments show that Transformers can learn to execute three representative recursions. We then train models from outcome or process rewards without oracle action or state targets, as in RL post-training. Different rewards yield distinct decision policies. Improving the alignment of process supervision with outcome-optimal action preferences shifts learned decisions accordingly. Such behavioral differences persist in a pretrained language model across textual paraphrases.

cs.AI

版本更新 · v2arXiv:2605.17552

Q-LocalAdam: Memory-Efficient Client-Side Adaptive Optimization for Edge Federated Learning

Vedant Waykole, Haroon R. Lone

原文摘要

On-device federated learning (FL) on edge devices is limited by the memory required by client-side adaptive optimizers. Maintaining optimizer states locally increases memory usage and can restrict the size of models that can be deployed on resource-constrained edge devices. State quantization can reduce this memory footprint, but uniform compression methods may perform poorly under skewed, non-IID data distributions because they can lose small-variance values that are important for stable optimization. We present \textbf{Q-LocalAdam}, a distribution-aware optimizer for memory-constrained edge devices. It applies block-wise linear quantization to the bounded momentum state and log-space quantization to the heavy-tailed variance state. A post-dequantization variance floor is used to maintain numerical stability. Q-LocalAdam reduces the persistent optimizer-state memory footprint while achieving convergence close to full-precision training. We evaluate Q-LocalAdam on physical edge devices and GPU, measuring persistent optimizer-state storage, process-level memory, and the latency introduced by quantization across human activity, vision, and language workloads. Our results show reduced peak resident memory with modest latency overhead, indicating that distribution-aware state compression can make adaptive optimization practical for edge deployments.

cs.LG

新论文 · v1arXiv:2611.22283

Towards Physics-Constrained Rectified Flow for Climate Downscaling: What Projections Can Certify?

Arun Sharma

原文摘要

Statistical downscaling restores the regional detail behind flooding and crop stress, but diffusion and flow-matching generators can produce sharp fields that break the conservation laws of their coarse input. Physics-Constrained Rectified Flow (PC-RF) is a conditional flow-matching downscaler. Its training penalties target wind divergence, negative precipitation, and mass drift from the coarse total, and a projection after sampling removes the last two exactly. On one corrupted synthetic field with no trained network, the projection drives the Non-negativity Violation Rate (NVR) from 0.236 and the Mass-Conservation Error (MCE) from 0.184 to zero, and the penalties lower the Divergence Error (DE) by about 30%. On ERA5 the projection gives PC-RF and RF-base, the penalty-free rectified-flow baseline, an NVR of exactly zero and an MCE of zero up to float32 rounding. The penalties add no perceptual gain that we can separate from training budget or run-to-run variation. At a matched budget on 2019-2021, PC-RF has no edge over RF-base (SSIM 0.939 vs. 0.942, PSNR 31.81 vs. 31.79, CRPS 0.771 vs. 0.772). Our unprojected CorrDiff reimplementation beats both flow models on PSNR (28.38) and CRPS (1.092) on 2020, yet 45% of its ensemble-mean precipitation cells are negative. On ERA5 a sign error in the divergence operator made that penalty and DE act on a deformation-like term. The fine reference is interpolated from ERA5 and accumulates precipitation over a longer window, so the perceptual scores cannot establish downscaling skill. NVR and MCE need one pass and no fine reference, so we argue they should run first as pass-or-fail checks, with DE as a diagnostic.

cs.AI

新论文 · v1arXiv:2611.22281

SeasonRainBench: Global Seasonal Precipitation

Qiuyan Sun, Mengxuan Chen, Haoyuan Liang, Guowen Li, Zhen Huang, Jianhao Lin, Juepeng Zheng, Haohuan Fu

原文摘要

Accurate seasonal precipitation forecasts help societies prepare for drought, flooding, and crop failure months in advance. While data-driven methods have made rapid progress in weather and subseasonal forecasting, operational seasonal precipitation prediction still relies on numerical ensemble systems. These systems are computationally expensive and exhibit biases and drift that increase with lead time, motivating more efficient data-driven alternatives. Yet this transition faces three linked challenges: comparing data-driven forecasts with operational ones requires (1) a global benchmark that does not exist, scoring them requires (2) metrics that capture operational priorities beyond anomaly correlation (ACC) and RMSE, and training them must (3) rely on a single, short observational record. To address the first two challenges, we introduce \BenchName, the first such benchmark, covering monthly precipitation forecasts 1--5 months ahead. It pairs global $1^\circ$ satellite precipitation with oceanic and atmospheric predictors, includes seven operational centers, and provides 310,944 simulated months from 52 climate models for training. Beyond standard scores, we introduce two operationally motivated metrics for anomaly strength and monthly rain-belt position. To address the third challenge, we develop \MODEL, which learns from simulations what the record lacks and transfers it to Earth through three modules: SimForced aligns the climate background, SimMemory transfers a history encoder pretrained on simulations, and SimAnalog supplements Earth labels with simulated analog outcomes. Extensive experiments show that \MODEL{} improves ACC, CRPS, anomaly strength, and rain-belt position by 12.0\%, 3.6\%, 2.9\%, and 0.3\% over the strongest operational or data-driven model, opening a path from simulated worlds to Earth system prediction.

cs.AI

新论文 · v1arXiv:2611.22280

XM-Sparse: High-Throughput Sparse-Attention LLM Serving over Scale-up Memory Pools

Yijun Sun, Xudong Liao, Zhuobin Huang, Kai Chen, Nikita Lazarev, Zhizhen Zhong

原文摘要

Sparse attention has been increasingly adopted by production LLMs for long-context inference, yet serving systems must still retain the full KV history, making GPU HBM capacity a key limit on serving concurrency. Existing systems offload the KV cache to host memory, but suffer from the limited bandwidth of PCIe and the limited capacity of host memory. Emerging scale-up memory pools (e.g., X-Mem) relieve both limits and open new opportunities for sparse-attention offloading. In this paper, we first conduct an in-depth analysis and reveal that, even with scale-up memory pools, decode throughput remains bounded by HBM capacity, as hot KV buffers and the fully resident indexer key (K-index) cache occupy substantial HBM. Moreover, the two caches exhibit distinct access patterns. Based on these insights, we present XM-Sparse, the first sparse-attention offloading system over scale-up memory expansion. The core idea of XM-Sparse is to offload both caches aggressively while matching each with an I/O mechanism tailored to its access pattern. Specifically, for dynamically selected KV entries, XM-Sparse fuses TMA-based retrieval with attention in a tile-level pipeline to hide on-demand retrieval latency. For predictable K-index accesses, it prefetches pages via SM-efficient GPU-initiated DMA and coordinates prefetching with latency-critical operations to avoid interfering with decoding. We have implemented XM-Sparse atop SGLang, and our experiments on DeepSeek-V3.2 and GLM-5.3 with an emulated scale-up memory pool show that XM-Sparse delivers up to 4.98$\times$ and 4.52$\times$ higher decode throughput than PCIe-based ECHO and HiSparse, respectively, and up to 1.34$\times$ over a naive integration on the same platform, across 16K-1M contexts.

cs.DC · cs.LG

版本更新 · v2arXiv:2611.01534

AbacusGS: Gaussian Splatting without Per-Splat Sorting

Jiacheng Ding, Yanliang Qi, Cong Guo, Peihua Mai

原文摘要

3D Gaussian Splatting sorts every splat by depth in every frame. Viewers that draw splats with the hardware rasterizer and sort them on the CPU, as most web and mobile viewers do, spend most of the frame time in this sort. We present AbacusGS, which stores the blending order of a 3DGS scene in the scene layout. AbacusGS groups the splats into octree boxes of at most 128 splats and stores each box in order along an axis chosen at write time. In each frame, AbacusGS sorts only the visible boxes, and one dot product per box decides whether the box is read forward or backward. A short order-aware fine-tuning converts an existing 3DGS model, and the over operator of 3DGS stays unchanged. On 13 standard scenes, AbacusGS loses 0.13 dB PSNR on average against an equally fine-tuned 3DGS model. In an OpenGL viewer that orders splats on the CPU, AbacusGS renders frames 4.5 times faster than per-splat sorting. We also present a compact format of 17 bytes per splat and a CUDA tile renderer that use the same order. The code will be released.

cs.CV · cs.GR

版本更新 · v2arXiv:2611.01467

See, Think, Use: Diagnosing Perception and Reasoning Failures in VLMs

Jiacheng Ding, Yanliang Qi, Cong Guo, Peihua Mai

原文摘要

We present See, Think, Use, a protocol that traces each answer of a vision-language model to perception, reasoning or the use of perception. The diagnosis shows a developer which stage to improve and directs test-time compute to the stage that fails. Benchmarks score the final answer, and an answer that rests on the image earns the same credit as a guess or a memorized prior. The protocol asks the model two extra questions per item and needs no access to its weights. A localization probe asks for a box around every instance and measures seeing, and a text-replacement probe answers from a description and measures thinking. Use is the rate at which the answer follows the model's own perception across counterfactual image pairs. The rate needs no answer labels, because the question and its prior are the same on both images. On two model families and six datasets the protocol shows that the knob that helps follows the task family. Answers follow the model's perception except on images that contradict a common prior, and asking the model to localize first repairs this failure when its boxes are right. Allocating compute by the measured stages comes close to the best uniform policy at lower cost and beats a random allocation of the same compute wherever a knob raises the short stage. We will release the code, the prompts and the per-question outputs.

cs.CV · cs.AI · cs.CL

版本更新 · v2arXiv:2611.16400

FuGIBench: Evaluating Reliable Preparation in Streaming Video

Changheng Cai, Zhichao Huang, Song Ren, Zhichen Zhao, Hongsen Liu, Chaoyang Zheng, Wenxi Wu, Yuan Fang 等 11 位作者

原文摘要

An assistant watching a worker can prepare for an emerging need before the worker is ready to receive help. As the need evolves, the assistant must maintain preparation until a response is due, or withdraw it if the need disappears. Video anticipation and interactive response benchmarks leave this interval between early evidence and response largely unmeasured. We introduce FuGIBench (Future-Grounded Interaction), a streaming-video benchmark in which a model chooses among waiting, reversible preparation, and response given a causal video prefix and a task contract. The benchmark contains 19,893 synthetic tool-assistance videos across ten tool tasks, in handover, ordinary-activity, and cancelled-need tracks. One primary score credits sustained preparation, incorporates response-interval alignment, and penalizes activity when the reference prescribes waiting. On a frozen set of 1,000 episodes, a reference system with a longer preparation lead has poorer response alignment, while similar handover leads accompany different levels of activity during ordinary work in matched scenarios. Together with cancellation analyses, these results show why evaluating preparation requires tracking decisions as needs develop, become actionable, or disappear.

cs.CV

新论文 · v1arXiv:2611.22278

DACE: Diversity-Driven Adversarial Co-Evolution for Robust LLM Safety Alignment

Peng Yu, Xiaoyu Wen, Zhida He, Ziyuan Zhou, Han Qi, Shao Zhang, Qiaosheng Zhang, Ying Wen 等 9 位作者

原文摘要

Ensuring robust safety alignment of large language models (LLMs) is increasingly difficult as adversarial attacks evolve and outpace alignment pipelines built on pre-collected data. Existing co-evolutionary frameworks let the defender chase a moving attacker, yet their training dynamics expose two compounding pathologies: \emph{attack strategy collapse}, where the attacker overfits to a narrow set of high-reward rewrites, and \emph{defense adversarial forgetting}, where the defender loses competence against earlier attacks as the attack distribution drifts. Existing remedies are partial: attacker-side diversity rewards score textual novelty rather than strategy novelty, and defender-side replay relies on discrete judge scores that ignore estimation uncertainty, response stochasticity, and defender non-stationarity. We introduce \textbf{\method}, a diversity-driven adversarial co-evolution framework that targets both pathologies jointly. On the attacker side, \method couples an explicit two-dimensional strategy space (risk category $\times$ attack style) with a \emph{normalized marginal coverage gain} reward, providing a bounded, non-vanishing exploration signal at the strategy layer. On the defender side, \method maintains a unified \emph{Bayesian adversarial replay pool} whose Beta--Bernoulli threat posteriors are refreshed by time decay and sampled via Thompson sampling, anchoring defender training to an evolving threat landscape. Across standard safety, automated-attacker, and general-capability benchmarks, \method improves robustness to out-of-distribution attacks while preserving general capabilities, and yields broader attacker strategy coverage.

cs.AI

版本更新 · v2arXiv:2611.18822

LongCat-Video-Avatar 1.6 Technical Report

Meituan LongCat Team, Xunliang Cai, Yiqi Chen, Feng Gao, Xiaoyu Kong, Le Li, Hongyu Liu, Xiaoming Wei 等 11 位作者

原文摘要

Recent advances in audio-driven avatar video generation have enabled increasingly realistic synthesis, yet achieving precise lip-audio synchronization, natural, human-like motion, and visual consistency over long sequences remains challenging. We present LongCat-Video-Avatar 1.6, which builds on LongCat-Video-Avatar 1.5 to improve expressive naturalness and reduce error accumulation during extended generation, e.g., exaggerated lip movements and overly red, temporally unstable lip coloration, enhance the naturalness of facial expressions and coordinated head and body movements. These improvements are driven by a refined post-training pipeline that integrates reinforcement learning, improved classifier-free guidance strategies for DMD-based few-step distillation, long-video rollouts, and on-policy distillation. Together, these refinements improve expressive realism and enable the generation of videos exceeding 10 minutes while maintaining visual quality and temporal consistency. Arena evaluations show that LongCat-Video-Avatar 1.6 outperforms its predecessor and competing commercial audio-driven avatar models, demonstrating the effectiveness of the improved post-training pipeline.

cs.CV