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

+1111新增论文
+874其中 AI 相关
+133已有论文修订
+97新增 AI 导读

arXiv 最近完成扫描:2026-11-20 07:46 · 全站最近入库:2026-11-19 19:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-19 07:52 — 2026-11-20 07:52(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:2026-11-20 07:46最近发现新增版本:2026-11-19 19:09库内论文最新更新时间:2026-11-19 02:59

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。

27,982 已收录论文23,249 AI 相关8,559 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 93 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-20 07:44(北京时间)
新论文 · v1arXiv:2611.24836

DiWALI:通过从人类和机器人游玩中推断潜在动作,在世界模型中学习扩散策略

DiWALI: Diffusion Policy Learning in World Models via Latent Action Inference from Human and Robot Play

Akshay L Chandra, Iman Nematollahi, Lingyue Zhao, Daria Kulikova, Angela P. Schoellig, Fabien Despinoy, Abhinav Valada

AI 速读 基于摘要

DiWALI 利用逆动力学模型从连续帧推断潜在动作,从而在无动作标签的人类和机器人游玩数据上训练世界模型,并基于潜在动作学习与改进扩散策略。

cs.RO

查看原文摘要

Learning robot manipulation is constrained by the availability of action-labeled robot data, for both imitation and world models. Humans can demonstrate skills in seconds, but their demonstrations lack the required robot actions. We introduce DiWALI, a framework that learns and improves diffusion policies within a world model trained on human and robot play. DiWALI infers latent actions from consecutive frames using an inverse dynamics model, enabling world model training without action labels, and grounds them in robot commands through a decoder trained separately on robot play. Skill policies are trained over latent actions from action-free human demonstrations and improved with reinforcement learning in imagination. On the CALVIN benchmark, DiWALI policies trained solely from human demonstrations substantially outperform baselines trained with robot demonstrations. Incorporating human play into the world model training also yields stronger policies after fine-tuning than using the same amount of robot play alone. Experiments on a physical robot show that skills learned primarily from human demonstrations are further improved in imagination with a world model trained on human and robot play. Together, these results demonstrate that human experience can support both manipulation skill acquisition and subsequent policy improvement, reducing reliance on action-labeled robot data. Videos, dataset and code are available at https://diwali.cs.uni-freiburg.de.

版本更新 · v2arXiv:2609.27917

When Does Reallocating a Sampling Budget Pay? Instance-wise Allocation for Neural Routing Solvers

Jinhyung Bae

原文摘要

Neural routing solvers sample many solutions per instance and keep the best, with the same number of samples for every instance. We measure how much a non-uniform split of the same total budget is worth and which property of a workload decides it. Best-of-k curves estimated by averaging random orderings of stored samples overstate the gain, and deciding on one half of the samples while evaluating on the other understates it; we use exact best-of-k curves, a U-statistic over all orderings. For three pretrained solvers (POMO, the Attention Model and Sym-NCO) on the traveling salesman problem, reallocation reduces the total optimality gap by at most 2.5% on the training distribution, so uniform allocation is close to optimal there. On workloads that mix in out-of-distribution instances it reduces the gap by 11.8-13.3% (1.1-1.8% of tour length) for the two solvers that degrade under the shift, and almost nothing for the one that does not. Across two heterogeneity mechanisms, three solvers, two confirmation collections and budgets of 25 to 200 samples per instance, the gain is predicted by the spread of difficulty within a workload, not its average. Allocating on a power-law curve fitted to twenty charged probe samples per instance recovers 88-93% of the uncharged optimum; successive halving does worse. A 2-opt local search, however, removes most of the gap at negligible cost, after which reallocation adds a few hundredths of a percent of tour length.

cs.LG · cs.AI · math.OC

新论文 · v1arXiv:2611.24641

Choosing the Right Representations: Task-Aware Fusion for Multimodal Materials Language Models

Huu Binh Ta, Alex Morehead, N. Benjamin Erichson, Ferdinando Fioretto, Sandeep Madireddy

原文摘要

Multimodal language models for materials science rely on effective mechanisms for incorporating three-dimensional structural representations. MatterChat addresses this challenge using a large bridge module between a pretrained structural encoder and the LLM, together with a separate contrastive alignment stage. We instead ask whether this bridge can adaptively select and combine representations from multiple structural encoders according to the target property. We find that supervised and generative material encoders provide complementary information for property prediction. We found that, for Zatom-1, this information is not concentrated in the final layer. Instead, the most informative representations occur at different depths depending on the target property. This motivates Task-Conditioned Attention Fusion (TCAF), a lightweight attention-based bridge that uses task-conditioned latent queries to combine CHGNet features with intermediate Zatom-1 representations. The bridge is trained directly on downstream prediction tasks, allowing representation selection and material-language alignment to be learned jointly without separate contrastive pretraining. Across classification and regression tasks, TCAF outperforms MatterChat, with up to 47.8% lower RMSE, while using substantially fewer trainable parameters and halving the overall training time. Further experiments incorporating UMA-S demonstrate TCAF's flexibility across encoder combinations, with the three-encoder configuration achieving the highest macro-average classification accuracy of 89.61%.

cs.LG · cs.AI

新论文 · v1arXiv:2611.23843

RePlan-VLA: Progress-Aware Replanning for Long-Horizon Vision-Language-Action Models

Apoorva Vashisth, Manav Kulshrestha, Damon Conover, Zachary Kingston, Aniket Bera

原文摘要

Vision-Language-Action (VLA) models have emerged as a promising paradigm for general-purpose robotic manipulation, yet their performance often degrades on long-horizon tasks that require multiple interdependent manipulations. Instead of burdening the VLA policies with long-horizon reasoning, memory, action generation, progress tracking, and error recovery, we argue for leveraging their short-horizon manipulation capabilities within an execution-aware framework. To this end, we introduce \textnormal{\textbf{RePlan-VLA}}, a VLA-agnostic framework that translates a VLA's short-horizon capabilities into reliable long-horizon execution. RePlan-VLA explicitly monitors the robot's motion for progress evaluation, and adaptively revises the future plan based on observed outcomes. We preserve the underlying VLA policy entirely, requiring no additional fine-tuning. Experiments on RoboCasa365 and CALVIN benchmarks show an improvement of 21.9% and 10.4% in success rate over the next strongest baseline. We also validate our approach for real-world deployment on a Franka Research 3 robot.

cs.RO · cs.AI · cs.LG

新论文 · v1arXiv:2611.23803

CoT-WAM: Collaborative Training of Vision-Language Models and World-Action Models via Plan Latents

Jinhao Zhang, Yushen Zuo, Ke Rui, Weitao Zhou, Haibao Yu, Honglong Tian, Xiaoheng Ma, Jie Mei 等 10 位作者

原文摘要

General robot manipulation requires both semantic competence and physical foresight. World action models (WAMs) acquire an understanding of and predictive capability over the physical world by jointly modeling visual foresight and action distributions. However, existing WAMs still lack semantic competence: they have limited ability to perform open-world scene understanding, semantic reasoning, and long-horizon planning. Although vision-language models (VLMs) provide these capabilities, effectively integrating their semantic priors into WAM prediction and control remains challenging. To address this challenge, we propose CoT-WAM, a jointly trained WAM-VLM framework. CoT-WAM uses a learnable embedding to exchange information between the VLM and the WAM, which is supervised by textual and goal-image signals during training to encode rich intent information. A tailored two-stage training strategy enables joint optimization and more effective collaboration between the VLM and WAM. We evaluate CoT-WAM on six benchmarks covering control performance, zero-shot generalization, and long-horizon planning. CoT-WAM achieves SOTA performance on LIBERO, CALVIN, and RoboCasa-Kitchen, and demonstrates stronger zero-shot generalization on LIBERO-Plus and LIBERO-Pro without fine-tuning. On RoboCasa365, it further improves long-horizon planning over baselines with and without external planners. Real-world experiments further demonstrate the effectiveness of our method in practical deployments.

cs.RO

新论文 · v1arXiv:2611.23746

ZERITH V2.5: From Predictive Power to Policy Learnability

Chenxi Han, Ruby, Epoch, Orion, Luke, Bin, Dusk, Lyra 等 21 位作者

原文摘要

When humans learn to manipulate an unfamiliar object, they may already be able to anticipate how its state will change while still being unable to control it skillfully. This gap between prediction and control motivates us to study how the priors acquired by video generation models can be used effectively by an action policy. We present ZERITH V2.5, a video-action model (VAM) in which the action policy reads intermediate features of a video generation model, and use it to study, under fixed demonstrations and a fixed total number of updates, what kind of predictive training improves subsequent action learning. We find that the way video predictive training is supervised affects subsequent action learning. In our experiments, adapting at many noise levels yields a large gain, while adapting only at the noise level used at deployment yields none. Weight-transfer experiments further show that the gain stays with the adapted weights when they remain in the video generation model and disappears when they are used only to initialize the action policy. Initial loss and gradient measurements at the start of joint training indicate that target-domain predictive training gives the video model a better starting state, making subsequent joint learning of prediction and action easier. Under matched data and training budgets, predictive adaptation improves control. We evaluate the complete policy on LIBERO, LIBERO-Plus, Meta-World MT50, and real-robot tasks.

cs.RO

新论文 · v1arXiv:2611.22977

A Common Exponential Lower Bound for Dantzig and Modified Simplex in Deterministic Discounted Games

Xinyu Qiao, Tao Yao

原文摘要

On a family of $N$-state deterministic two-player discounted games, every run that chooses adaptively between Dantzig's rule and modified simplex requires $2^{Ω(N)}$ switches from a specified initial strategy, regardless of tie breaking. More generally, a return-gate construction lets Dantzig reproduce any finite Bland prefix in an undiscounted deterministic Markov decision process with rational rewards whenever every visited policy reaches a common sink from all states. Applied to an acyclic binary counter, deleting the minimizer's return actions preserves every fixed maximizing strategy's value at each original state while reducing Dantzig's switch count from exponential to linear. With returns, halving the total optimality gap can take $Ω(h/\log h)$ switches at discount horizon $h=(1-γ)^{-1}$, even on games solvable in linear arithmetic time. Conversely, Dantzig is strongly polynomial when strongly connected components have bounded size, each state has a bounded number of actions, and the condensation graph has height $O(\log N)$.

cs.GT

新论文 · v1arXiv:2611.22388

Hidden Variables Behind Real-Robot Data Quality: Lessons from Hundreds of Thousands of Hours

Maniformer Real-World Manipulation Team, Che Tsang Calvin, Xingyu Chen, Bowen Jiang, Naixiang Gao, Xiongfeng Cai, Enlin Gu, Ruiying Li 等 42 位作者

原文摘要

As real-robot data collection scales, collection-system design shapes both operator efficiency and the behaviors learned by robot policies. Workspace limitations, control-interface friction, and latency can introduce pauses, detours, and corrective motions into otherwise successful demonstrations. We investigate these effects across embodiment, teleoperation control, and data consistency, combining a kinematic study of 25 dual-arm robots, controlled teleoperation experiments, and real-robot evaluation of fine-tuned $π_{0.5}$ policies. The embodiment study shows that positional reachability overstates orientation-aware task coverage and that modest shoulder motion can substantially expand the effective workspace. Teleoperation experiments show that parallelizing body-part control reduces task completion time by 43-53%, while compliant control improves success and reduces peak contact forces on constrained and contact-rich tasks. Policy-learning experiments show that demonstrations must capture the visual information underlying action choices: active-search demonstrations improve arm-selection accuracy from chance (50%) to 100%. Calibration experiments reveal that mixed-calibration training data remain usable in our tested configurations when an inference-aligned subset is retained, whereas inference-time miscalibration causes all 30 trials to fail. Policies also retain approximately 90% of the extra path length in detour-contaminated demonstrations, showing how inefficient operator motions persist in learned behavior. Together, these findings connect collection-system design to demonstration quality and downstream policy execution. They motivate evaluating effective workspace, control efficiency, observation completeness, calibration alignment, and motion quality before scaling collection, and addressing system-induced compensations at their source.

cs.RO

版本更新 · v2arXiv:2605.14434

Efficient Generative Retrieval for E-commerce Search with Semantic Cluster IDs and Expert-Guided RL

Jianbo Zhu, Xing Fang, Jing Wang, Mingmin Jin, Bokang Wang, Guangxin Song, Zhenyu Xie, Junjie Bai

原文摘要

Generative retrieval offers a promising alternative by unifying the fragmented multi-stage retrieval process into a single end-to-end model. However, its practical adoption in industrial e-commerce search remains challenging, given the massive and dynamic product catalogs, strict latency requirements, and the need to align retrieval with downstream ranking goals. In this work, we propose a retrieval framework tailored for real-world recall scenarios, positioning generative retrieval as a recall-stage supplement rather than an end-to-end replacement. Our method, CQ-SID (Category-and-Query constrained Semantic ID), employs category-aware and query-item contrastive learning along with Residual Quantized VAEs to encode items into hierarchical semantic cluster identifiers, significantly reducing beam search complexity. Additionally, we develop EG-GRPO (Expert-Guided Group Relative Policy Optimization), a reinforcement learning approach that aligns generative recall with downstream ranking under sparse rewards by injecting ground-truth samples to stabilize training. Offline experiments on TmallAPP search logs show that CQ-SID achieves up to 26.76% and 11.11% relative gains in semantic and personalized click hitrate over RQ-VAE baselines, while halving beam search size. EG-GRPO further improves multi-objective performance. Online A/B tests confirm gains in GMV (+1.15%) and UCTCVR (+0.40%). The generative recall channel now contributes substantially in production, accounting for over 50.25% of exposures, 58.96% of clicks, and 72.63% of purchases, demonstrating a viable path for deploying generative retrieval in real-world e-commerce systems.

cs.IR · cs.AI

新论文 · v1arXiv:2611.21515

超越排行榜名次:预算约束下的软件工程智能体选择

Beyond Leaderboard Ranks: Selecting Software Engineering Agents Under Budget Constraints

Srinath Srinivasan, Tim Menzies

AI 速读 基于摘要

作者利用HAL中58,197条运行记录,比较在显式评估预算与部署成本上限下选择软件工程智能体的策略,指出排行榜名次证据薄弱,并报告逐次减半在已知任务成本时显著节省评估开销。

cs.SE

查看原文摘要

Teams that adopt LLM agents for issue resolution must decide which agent to pay for, and how much to spend finding out. Leaderboard ranks give weak evidence for this decision. Accuracy differences may be statistically unresolved, reported costs may be inconsistent, and evaluating candidate agents can itself be expensive. We investigate budget-constrained agent selection using 58,197 run--task records extracted from the Holistic Agent Leaderboard (HAL), focusing on SWE-bench and six additional benchmarks. Across seven benchmarks, 27 of 35 steps along the accuracy--cost Pareto frontiers are statistically unresolved. On SWE-bench, a run costing $34.16 trails the $170.78 leader by 14.6 percentage points, but the available tasks cannot statistically distinguish their accuracy. Reported costs also differ from standardized token-based estimates by up to 11.7x, while selective task evaluation can bias estimates of full-benchmark execution cost. We therefore compare evaluation strategies under explicit evaluation budgets and deployment-cost caps. Under advance knowledge of task costs, successive halving reaches our uncertainty-aware selection target on SWE-bench using 6.6% of exhaustive evaluation spend, achieving the lowest cost on six of seven benchmarks and tying on the seventh. Its advantage narrows when task costs are unknown before execution or exact-best selection is required. Our findings motivate decision-oriented agent evaluation that accounts for accuracy uncertainty, validates execution costs, and compares complex selection methods against simple adaptive allocation.

新论文 · v1arXiv:2611.21249

Heteroscedastic Gaussian Process Q-learning for Sample Efficient Experimentation

Maximilian Bloor, Max Mowbray, Calvin Tsay, Ehecatl Antonio del Rio Chanona

原文摘要

Self-driving laboratories must learn to control a process from tens to hundreds of expensive experiments, too few for deep reinforcement learning (RL). We learn the Q-function directly with a Gaussian process (GP), which avoids learning the dynamics, and show that the GP posterior supplies the learning rate of Q-learning. If each Bellman target is treated as a noisy measurement of the Q-function, the learning rate that minimises the mean squared error is the Kalman gain, and applying it to every observed transition is equivalent to GP regression. Heteroscedastic Gaussian Process Q-learning (HGPQ) sets the noise variance of each target with a heteroscedastic GP, which learns how unreliable the targets are across the states and inputs. Targets from noisy operating conditions, or from regions where the value function has large gradients, then move the estimate less than targets from low-noise regions, so HGPQ learns quickly where the data are reliable without being misled where they are not. For kernel value iteration with per-target variances that are assigned when a target is collected, we prove a high-probability bound on the error of the value estimate, uniform over the explored region. Pooled over five simulated benchmarks, which include a chemical reactor and a bioprocess, HGPQ approaches the best return found by any method in fewer episodes than GP, kernel and deep RL baselines.

cs.LG · eess.SY

新论文 · v1arXiv:2611.21020

SWE-OPD: On-Policy Distillation for Agentic Coding

Yuhang He, Yanyu Chen, Chuanzhe Guo, Jie Zhu, Zichen Tian, Hongyu Ge, Zixuan Ren, Huang He 等 14 位作者

原文摘要

In this work, we study on-policy distillation (OPD) for transferring agentic coding capabilities from a large teacher model to a smaller student model. Prior work has focused primarily on reasoning and short-horizon agentic tasks. We present SWE-OPD, a minimal OPD study for software engineering, where the student interacts with executable repositories and the teacher supervises the prefixes visited by the student. Across three teacher--student pairs, two training sources, and three SWE benchmarks, we find that on-policy supervision transfers coding capability: supervision from Qwen3.6-27B increases the SWE-bench Pro resolve rate of Qwen3.5-9B from 35.9% to 39.4%. Refreshing trajectories is not strictly required either, as an offline variant with cached teacher scores still improves over the student baselines while cutting end-to-end GPU-hours by 59.4%--72.6% and halving peak GPU allocation, at benchmark-dependent accuracy. Furthermore, we extend the offline variant to corss-tokenizer scenarios: distilling GLM-4.5-Air from DeepSeek-V4-Flash-0731 raises its SWE-bench Pro and Multilingual resolve rate from 35.8% to 39.7% and from 41.8% to 46.2%, around 4% and 4.6% improvement, respectively. We also find that a decreasing loss, however, does not necessarily indicate improved agent performance: under top-$K$ reverse-KL loss, one student exhibits severe repetition despite a decreasing loss. We identify a support-mass gradient term as a potential mechanism underlying this failure.

cs.AI · cs.LG

版本更新 · v2arXiv:2611.16861

Visual Cipher Bench: Benchmarking LLM Agents on Long-Horizon Visual Reasoning

Deyao Hong, Jiayi Jiang, Yuchun Chen, Yixuan Shi, Wenyi Li, Zheng Jiang, Chengxing Xie, Mingyu Su 等 15 位作者

原文摘要

While language-model agents excel at long-horizon coding and terminal tasks, this success relies on two conditions: explicit text specifications and immediate execution feedback. However, neither condition holds in real-world problems where critical information exists only in images, leaving it unclear whether long-horizon capability transfers from text to visual reasoning. In such visual settings, because rules and clues resist text transcription and lack test-case feedback, agents must repeatedly inspect images, simulate spatial transformations, and autonomously cross-check clues. Despite this challenge, existing evaluations fail to capture such capability: visual reasoning benchmarks largely reduce tasks to single-turn, short-chain question answering, whereas agent benchmarks focus on text-based environments or interactive states that permit trial-and-error. To bridge this gap, we introduce Visual Cipher Bench, a collection of 40 original visual puzzles across four categories: spatial manipulation and simulation, logical constraints and planning, rule induction, and knowledge-based decoding. Each puzzle provides only one to three black-and-white line drawings and a minimal submission instruction, requiring agents to discover the underlying solving mechanism entirely from the images. Across 29 reasoning-effort configurations of eight frontier models, even the strongest model, GPT-6 Astra (max), achieves only 26.7% accuracy, while five of the eight models score below 1%. Yet on a ten-task subset, providing a single non-spoiler mechanism hint boosts accuracy from 12.2% to 51.1%. This sharp contrast shows that the primary bottleneck of current agents lies not in executing multi-step plans, but in autonomously discovering rules from visual evidence. Visual Cipher Bench thus provides a rigorous benchmark for measuring long-horizon visual reasoning in agents.

cs.CV

新论文 · v1arXiv:2611.20010

PlanDiv: Test-Time Scaling with Latent Plans for Vision-Language-Action Models

Hanbo Zhang, Lianlei Shan, Liuhaichen Yang, Zezhi Tang

原文摘要

Test-time scaling improves vision-language-action (VLA) policies by drawing several action candidates and executing the best one. We find that its returns diminish quickly when the candidates are resampled from one decoder: they differ by noise rather than by plan, and for each of three resampling sources, going from four to eight candidates adds only 0.06-0.07 to the average number of completed tasks on CALVIN. We introduce PlanDiv, an action head that makes the candidates differ by plan: a lightweight planner proposes K latent plan hypotheses, a shared masked-diffusion executor decodes one action chunk per plan from a single backbone pass, and the policy executes the chunk that its own likelihood ranks highest. The plans are learned from demonstrations alone, without plan labels or rewards. With the backbone, data and training recipe held fixed, eight plan-conditioned candidates raise five-task chain success on CALVIN ABC$\rightarrow$D from 56.9% to 69.9%, while eight candidates from autoregressive, plan-free or same-plan resampling reach at most 64.9%. The same-plan baseline is PlanDiv's own checkpoint with the plan held fixed, so this gap comes from the plans, not from drawing more samples. On LIBERO-Long, four plans remove 40% of single-candidate failures, against at most 21% for matched resampling. On a real Franka arm the gain appears where a task admits more than one solution: PlanDiv is on par with a single-output OpenVLA-OFT+ baseline on a one-solution task and succeeds in 51 of 54 trials against 40 of 54 on two tasks with two solutions.

cs.RO · cs.CV · cs.LG

新论文 · v1arXiv:2611.19698

Data Attribution Should Steer Pretraining

Glenn Matlin, Chandreyi Chakraborty, Mohammed Rehan Parwani, Saehee Eom, Anthony Wen-Ming Zang, Akhil Theerthala, Arjun Chatterjee, Tejas Khandwekar 等 11 位作者

原文摘要

Pretraining data curation decides what a model learns from, yet what makes a document worth learning from depends on what the model already knows. Influence-driven selectors already rank documents to improve one run; we argue that training-data attribution should steer pretraining by identifying testable data properties that become reusable curation criteria, evaluated on new data against cheaper selectors at matched total cost, discovery included. A document ranking is consumed by the run it was computed for, whereas an explained data property can be tested across corpora and checkpoints, especially where familiar quality signals diverge from measured usefulness. Because historical contribution, an understandable explanation, and future data value require distinct evidence, we specify a feedback loop that interprets attributed sources, tests candidate data changes, and states for five claim types what evidence each requires and what none establishes. We close with a falsifiable comparison against perplexity, classifier, and mixing baselines at equal total compute.

cs.LG · cs.AI · cs.CL

版本更新 · v3arXiv:2605.00087

DeGenTWeb: A First Look at MGT-dominant Websites

Sichang Steven He, Calvin Ardi, Ramesh Govindan, Harsha V. Madhyastha

原文摘要

Many news reports have claimed that machine-generated text (MGT) from large language models (LLMs) is taking over the web. However, these claims are typically not based on a representative sample of the web, and the methodology underlying them is often opaque. Consequently, we lack an understanding of the true prevalence and characteristics of MGT on the web. We describe DeGenTWeb, which systematically identifies MGT-dominant websites: sites whose informational prose content is mostly MGT, with little human input. We show how to adapt detectors of MGT for use on web pages, and how to aggregate detection results from multiple pages on a site for accurate site-level categorization. DeGenTWeb mislabels only 0.12% of human-authored sites as MGT-dominant. We find that 20.9% of the sites archived by Common Crawl between January and July 2026 and 13.6% of the sites linked from Bing's search results in 2025 are MGT-dominant. These figures are lower bounds on the prevalence of MGT-dominant sites. We also find that the fraction of MGT-dominant sites is growing over time, and that many such sites share the same owner. We will release all of our code and data after publication.

cs.NI · cs.AI · cs.CY · cs.IR · cs.LG

新论文 · v1arXiv:2611.19279

Rethinking KV Caching for LLM Decode: Free-Space Optical KV Regeneration for Reduction of Memory Traffic

Aritra Bagchi, Bin Gao, James Spall, Jakub Pietrak, Weng-Fai Wong, Tulika Mitra

原文摘要

Long-context large language model (LLM) decoding repeatedly reads the historical key-value (KV) cache, making multi-head attention (MHA) memory-bandwidth-bound. Caching layer-input activations instead halves historical-cache storage and traffic, but requires regenerating KV from each cached activation. On digital accelerators such as GPUs, this regeneration takes far longer than fetching the activation from HBM. The KV regeneration overhead outweighs the savings from reduced memory traffic, shifting decode attention from a memory bottleneck to a compute bottleneck. We present OptiDecode, a hybrid optoelectronic decode-attention accelerator that regenerates historical KV using free-space optics and computes attention digitally. Projection weights remain resident on optical tiles, exploiting reuse across historical tokens for high-throughput optical regeneration. Scaling to large models introduces two challenges. First, distributing weights across tiles requires sending the same activations to every head group, while each readout die produces only partial KV. Activation replay shares one set of electrical-to-optical converters across head groups without costly broadcast. In-situ attention avoids transferring partial KV from readout dies to reducers by computing on them locally and exchanging only partial scores, probabilities, and partial outputs. Second, processing an activation chunk through all dependent stages takes longer than one activation-fetch interval. Chunk pipelining overlaps successive chunks so that OptiDecode sustains the HBM activation-fetch rate. OptiDecode demonstrates 1.92-1.99x decode-attention speedup over an ideal GPU KV-read baseline at equal peak HBM bandwidth (A100-B200), for 32K-token contexts across evaluated 30B-175B MHA models, approaching the 2x upper bound from halving historical memory traffic.

cs.AR · cs.ET

版本更新 · v2arXiv:2608.07314

TEMPO:面向视觉-语言-动作模型的语义-动作解耦强化学习后训练

TEMPO: Semantic-Action Decoupled RL Post-Training for Vision-Language-Action Models

Ziheng Liu, Quantao Yang

AI 速读 基于摘要

TEMPO 提出一种语义与动作解耦的双时间尺度强化学习后训练框架,冻结预训练视觉-语言主干,分别以不同频率更新语义投影层和动作专家,以缓解潜在动作漂移并提升策略性能。

cs.RO · cs.CV

查看原文摘要

Vision-language-action (VLA) models are commonly adapted to downstream manipulation tasks via supervised fine-tuning (SFT) or online reinforcement learning(RL) post-training. SFT is prone to distribution mismatch, and existing RL approaches typically optimize all trainable components within a single RL loop at a uniform update rate, although these components play distinct functional roles. We propose TEMPO, a semantic-action decoupled, two-timescale RL post-training framework for VLA models. TEMPO freezes the pretrained vision-language backbone and adapts two components with dedicated RL loops: the semantic projection layer, which produces the latent action conditioning the action expert, and the low-level action expert itself. The semantic projection layer is updated infrequently and the action expert frequently, which aims to limit latent-action drift while allowing the action expert to learn from online feedback. Our analyses show that, compared with equal update frequencies, less frequent semantic updates reduce latent-action drift and improve policy performance. Experiments on the CALVIN benchmark and real-world manipulation tasks demonstrate that TEMPO outperforms both pretrained state-of-the-art VLA models and the RL post-training baseline, while reaching and maintaining higher evaluation rewards on three real-world tasks.

新论文 · v1arXiv:2611.16759

Propositional Dynamic Logic without tests does not have Craig Interpolation

Malvin Gattinger

原文摘要

It was shown recently by Borzechowski et al. (2026) that Propositional Dynamic Logic (PDL) has Craig Interpolation. Whether test-free PDL, i.e. PDL without the `?' operator, has interpolation remained open. Here we answer this question in the negative with a counterexample: a test-free PDL validity for which no test-free interpolant exists. The validity only uses one atomic program and the Kleene star, so it implies that the basic modal logic K extended with the reflexive transitive closure does not have interpolation, as previously shown by Studer (2009). The example was originally obtained using LLMs and Lean. We describe our methodology and provide links to all prompts, LLM output summaries as well as the Lean proof.

cs.LO · math.LO

新论文 · v1arXiv:2611.16662

MuSeToM: Multi-Sensory Evaluation of Belief Revision with Textual Cues Across Six Senses

Chunkit Chan, Kawai Chung, Sze Heng Douglas Kwok, Yuxuan Liu, Yauwai Yim, Tianshi Zheng, Haochen Shi, Qing Zong 等 16 位作者

原文摘要

Theory of Mind (ToM), the ability to infer mental states in others, is pivotal for human social cognition. Existing evaluations of ToM in large language models (LLMs) mainly focus on static belief attribution and false-belief reasoning, leaving a critical aspect of social reasoning underexplored: can LLMs decide whether perceptual evidence is sufficient to justify another person's belief revision? To address this gap, we present MuSeToM, a human-validated, text-based benchmark for evaluating epistemic belief revision in ToM reasoning. MuSeToM contains 2,415 instances adapted from BigToM and OpenToM, with textual cues spanning six sensory categories, and tests whether models can determine when a protagonist should revise a belief and infer the resulting updated belief state. Using MuSeToM, we systematically evaluate state-of-the-art LLMs under direct and chain-of-thought prompting, revealing a pronounced dissonance: while models often detect that a belief may change, they struggle to identify the updated belief and frequently over-update on perceptually salient but epistemically insufficient cues.

cs.CL · cs.AI

版本更新 · v3arXiv:2610.30022

Density Constraints Overcome the Energy-Density Trade-Off in Exchange-Correlation Functionals

Yuhao Chen, Donald G. Truhlar, Xiao He

原文摘要

Exchange--correlation functionals are trained mainly on energies, and their densities often degrade as their energies improve. We present a supervised-learning algorithm that trains functionals on energies and accurate densities without self-consistent-field (SCF) solutions. From one Fock build at a fixed accurate density, it predicts the density that the functional would reach self-consistently and penalizes its deviation from the reference; for functionals linear in their parameters, training is convex. Refitting the MN15 form to 1,500 reaction energies under a constraint on 98 coupled-cluster densities gives the density-constrained hybrid DCH-D4. On 19 held-out energy sets its mean normalized error ratio is 0.96, against 1.76 for MN15, and on 145 densities outside the fit its mean density error is 22 times lower than that of MN15. Deep-learning neural GGA and meta-GGA functionals (NN-GGA and NN-mGGA) trained on only 6 energies and 11 densities lower the atomization-energy error on 103 molecules from 14.6 to 5.6~kcal/mol relative to PBE and from 3.7 to 3.3~kcal/mol relative to SCAN, while roughly halving the density error of each base functional. Accurate densities can thus enter functional training at the scale of modern energy databases.

physics.chem-ph · cs.AI

新论文 · v1arXiv:2611.16363

Vector-Space Projection and Unified Complex-Valued Acceleration for Scaling Matrix-Based Brain-Inspired Representations

William Youngwoo Chung, Hyunwoo Oh, Calvin Yeung, Hansen Jin Lillemark, Hamza Errahmouni Barkam, Mohsen Imani

原文摘要

Symbolic reasoning at scale requires representations that are both expressive and hardware-efficient. Hyperdimensional Computing (HDC) provides a promising brain-inspired framework, encoding concepts as high-dimensional vectors composed through efficient algebraic operations. However, the widely used Fourier Holographic Reduced Representation (FHRR) relies on commutative binding, which cannot encode order or directionality. Prior FPGA/ASIC accelerators recover this through costly permutation logic, incurring irregular memory access and high control overhead. Generalized Holographic Reduced Representations (GHRR) instead redefine binding as non-commutative matrix multiplication, naturally encoding sequences, graphs, and hierarchies without permutations-but at the cost of turning every element into a matrix, introducing compute and storage overhead that limits scalability. In this work, we propose a flattening methodology that preserves GHRR's matrix-based encoding while executing training and inference directly in vector space, functionally equivalent to FHRR inference yet free of permutation logic. We further design a custom 28 nm ASIC that fuses binding and similarity into a unified complex-valued datapath, with dual-DMA streaming and runtime normalization for accurate inference. Against a PyTorch baseline on an NVIDIA RTX 4090 GPU, the prototype delivers 1.36x-1.56x higher throughput and 16.2x-18.6x better energy efficiency.

cs.AR

新论文 · v1arXiv:2611.15154

SciSelect-Gym: An Environment and Benchmark for Active Learning and Bayesian Optimization Configuration Selection

Ying-Peng Tang, Tianle Zhang, Wanlong Fang, Ng Jing Xiang, Alvin Chan

原文摘要

In many scientific tasks, scientists repeatedly choose a small set of candidates to test, learn from the results, and use the observations to guide the next selection. Methods based on active learning~(AL) or Bayesian optimization~(BO) have been proposed for these iterative screening campaigns with a surrogate model. Yet, selecting the surrogate model and acquisition function for a new dataset remains challenging due to limited empirical evidence, while the effectiveness of large language models (LLMs) in making such decisions remains unclear. To address these gaps, we introduce SciSelect-Gym, an environment for simulating AL/BO data selection campaigns and evaluating configuration choices. We conducted simulations of AL/BO data selection campaigns across several settings and more than 100 molecular datasets. Using completed simulation campaigns as context improves LLM recommendation shortlists on held-out datasets more consistently than it improves the quality of their first choice. SciSelect-Gym provides a simulation platform and recommendation protocols for learning how to choose configurations before an acquisition campaign. An implemented agent use case connects these campaign records and simulation tools to LLM-generated configuration-selection policies. The code is provided as supplementary material.

cs.CE

版本更新 · v2arXiv:2610.37914

Muon 使首个混合专家层更难平衡

Muon Makes the First Mixture-of-Experts Layer Harder to Balance

Marcos Hernanz

AI 速读 基于摘要

作者报告在 55M 与 108M 激活参数模型中,Muon 相比 AdamW 显著放大首个 MoE 层路由偏置跨度,导致该层负载更难平衡。

cs.LG · cs.CL

查看原文摘要

Many open mixture-of-experts (MoE) language models train with Muon instead of AdamW. Most balance the load with a bias on each expert's routing score, adjusted after every step. Their recipes disagree on whether Muon should also train the router. In models with 55M and 108M active parameters, we measure how the optimizer changes the span of these biases, the largest minus the smallest in a layer. The span grows when all tokens favor the same experts, so it shows how hard a layer is to balance. At learning rates tuned on shorter runs, Muon raised the span of the first MoE layer at every seed, roughly doubling it on average, while the other layers changed much less. The rise came from Muon on the rest of the network, whichever optimizer trained the router. In those runs, the inputs to the first router shared a direction, and soon after the warmup its busiest expert averaged four to seven times its fair share. Over the first 3,000 steps, subtracting the batch mean from these inputs removed this overload while barely changing the loss. On the smaller model, halving Muon's learning rate removed most of the rise and slightly improved the loss. Moving only the router to AdamW made the loads shift more during the warmup and slightly raised the loss, so at this scale the results favor keeping the router on Muon.

版本更新 · v2arXiv:2610.35495

Query Complexity of Learning Axis-Aligned Partitions

Hadley Black, Arya Mazumdar, Barna Saha, Geelon So

原文摘要

This work establishes the query complexity of learning high-dimensional boxes in the exact query learning setting. Let $X \subset \mathbb{R}^d$ be a set of $n$ points, labeled by a collection of $k$ boxes: the label of each point in $X$ is determined by the unique box containing it. A learner aims to exactly recover the labels for every point in $X$ given access to a label oracle. We consider three increasingly-structured settings. In the most general setting, there are no conditions on the boxes, and they may overlap. We show that CAL (Cohn-Atlas-Ladner 1994), a classical active learning algorithm, is minimax-optimal with query complexity $Θ(kd \log (n/kd))$. When the boxes are disjoint, we examine two regimes. In the large-$k$ regime, CAL remains minimax-optimal. For the constant-$k$ regime, an optimized variant of the Halving algorithm (Hegedüs 1995) can beat CAL by a $\log(d)$ factor. Finally, in the most-structured settings, the boxes correspond to the leaves of a constant-sized decision tree. We prove a tight query complexity of $Θ(d + \log n)$ that is additive in $d$ and $\log n$, generalizing a recent result (Black-Larsen-Mazumdar-Saha-So 2026) that establishes the query complexity of learning decision stumps.

cs.DS · cs.CG · cs.LG

新论文 · v1arXiv:2611.13425

Prompt2Echo (P2Echo): Text-Prompted Multi-View Echocardiography Segmentation with One Model

Moein Heidari, Alvin Kimbowa, Mahdi Tabatabaei, Afshin Bozorgpour, Sina Ghorbani Kolahi, River Jiang, Wenjin Chen, David J. Foran 等 11 位作者

原文摘要

Automated segmentation in multi-view echocardiography is essential for comprehensive cardiac assessment, yet remains challenging due to substantial anatomical and appearance variability across standard acoustic windows. In this paper, we present P2Echo, a unified framework capable of segmenting diverse cardiac structures across multiple standard echocardiographic views using a single model controlled by structured view-aware text prompts. Unlike recent foundation model adaptations that rely on implicit latent priors, P2Echo explicitly conditions the segmentation process via a novel Dynamic Image-Text Alignment (DyITA) module. DyITA integrates a differentiable routing mechanism and token differential operators to dynamically modulate decoder features according to the requested view and target structure. This allows the network to adaptively switch its processing pathways for different views without architectural changes. Extensive experiments on a large-scale internal dataset (seen during training) and held-out evaluation on three external datasets demonstrate that P2Echo matches or exceeds both view-specific specialists and state-of-the-art universal models at a fraction of their computational cost, offering an efficient, controllable solution for multi-plane cardiac analysis. Code is available at: https://github.com/moeinheidari7829/P2Echo.

cs.CV

版本更新 · v2arXiv:2610.41927

Variance Reduction for Decentralized Stochastic Optimization with Multiple Local Updates

Jie Liu, Nicola Bastianello, Wenying Xu, Daniel W. C. Ho, Karl H. Johansson

原文摘要

Multiple local updates are widely used in decentralized optimization to enhance communication efficiency, often in conjunction with stochastic gradients when exact gradient computation is costly. However, such algorithms generally converge only to a neighborhood of the optimum, whose size depends on the stepsize and the variance of gradient noise. Variance reduction has been incorporated into ADMM-based methods to achieve exact convergence in expectation, but their stepsizes typically scale inversely with both the square root of the network size and the cube of the number of local updates, leading to slow convergence as the network size or the number of local updates increases. In this paper, we propose a novel primal-dual-based decentralized optimization algorithm that integrates variance reduction with multiple local updates, with a stepsize independent of the network size and inversely proportional to the number of local updates (rather than its cube), thereby enabling significantly larger stepsizes and faster convergence. Moreover, to the best of our knowledge, we establish the first linear convergence guarantee in expectation for decentralized optimization with multiple local updates under stochastic gradients in the strongly convex setting. Beyond the theoretical guarantee, the proposed method exchanges only one variable per round, halving the communication cost compared with gradient-tracking-based methods. Numerical experiments further demonstrate its advantages in convergence speed and accuracy.

math.OC · cs.LG

新论文 · v1arXiv:2611.12856

WorkflowArena:面向基于大语言模型的多智能体系统的可执行工作流编排基准

WorkflowArena: Benchmarking Executable Workflow Orchestration in LLM-Based Multi-Agent Systems

Zheng Jiang, Fang Niu, Bo Peng, Houde Qian, Wenyi Li, Deyao Hong, Mingju Gao, Yizhe Chi 等 13 位作者

AI 速读 基于摘要

提出 WorkflowArena 基准,用 145 个长周期任务评估大模型多智能体系统自主编写并执行工作流编排的能力。

cs.CL · cs.SE

查看原文摘要

Large language model agents are increasingly deployed as multi-agent systems rather than as a standalone model operating within a single context. Recent agent harnesses allow the model to instantiate worker agents and to author the executable workflow that coordinates them, shifting the design of system structure from developers to the model at run time. The quality of this organization therefore constitutes a distinct capability, yet existing benchmarks either measure end-to-end task completion or evaluate coordination under structures specified in advance. We formalize this capability as workflow orchestration and introduce WorkflowArena, a benchmark of 145 long-horizon tasks spanning 8 professional domains, from software engineering and formal mathematics to legal analysis and 3D production. Each task specifies an initial workspace and an integrated deliverable composed of interdependent components, which is assessed by a hidden verifier. Agents determine autonomously whether and how to delegate, and all delegation is recorded, enabling joint evaluation of deliverable quality and the organization of the underlying work. Experiments with nine models show that no single configuration performs best across all domains and that systems exhibit distinct delegation strategies: at high effort, Claude models in Claude Code invoke workflows less frequently but with larger worker groups, whereas GPT models in Codex delegate more frequently in smaller units. When the executing model is held fixed on a challenging subset of tasks, workflows authored by Claude Opus 5.5 substantially improve weaker executors, indicating that workflow organization is itself a measurable determinant of task performance.

新论文 · v1arXiv:2611.12122

System Optimizations for Cost-Efficient Agent Harnesses

Jongseok Park, Jiaxiang Yu, Eric Zhou, Tony Hong, Sunga Kim, Yusuf Afifi, Ian Wong, Ion Stoica 等 9 位作者

原文摘要

AI agent harnesses play the critical role of providing LLMs with an interface to a computing environment. Recent studies show that AI agents' task-execution accuracy depends mostly on the LLM used, while harness-side variances such as system prompts or tools have minimal impact. Rather, such harness-side variances show a larger impact on the cost of agent executions, achieving cost reductions up to 2.6 times on the same model and task. To further explore these harness-side cost optimization opportunities, we analyze the cost accumulation patterns across various agent traces and identify system opportunities to reduce token use and LLM calls without modifying the model or affecting task accuracy. Based on this analysis, we design Agency, a harness interposition platform that provides a unified abstraction and allows easy and portable implementation of harness-side system optimizations on existing off-the-shelf harnesses such as Claude Code or OpenAI Codex. Using Agency, we implement proof-of-concept optimization techniques based on our analysis and achieve cost reductions up to 35% across various real-world benchmarks on existing harnesses without compromising accuracy.

cs.AI

新论文 · v1arXiv:2611.11674

Brain-Aligned Latent Actions: Imposing Premotor Cortex Geometry on a Robot's Action Vocabulary

Amit Jadhav

原文摘要

Robot policies increasingly act through a small vocabulary of discrete latent actions learned from unlabeled video, yet training leaves open which motion receives which code. We find this choice highly unstable: two independently trained runs of Moto's latent motion tokenizer assign the same robot clips to nearly unrelated codes (adjusted Rand index 0.001), and evaluating one tokenizer in bf16 instead of fp32 reassigns 74% of its tokens. We use this freedom to impose the representational geometry of macaque ventral premotor cortex (PMv), recorded while a monkey drew shapes, on the tokenizer's code assignment, with a straight-through similarity loss trained alongside its reconstruction objective on robot video. In one training run per target, on five shape categories the loss never saw, the PMv-trained tokenizer reaches a representational similarity of 0.52 with PMv, against 0.20-0.26 for released and no-prior tokenizers and 0.42 for the 95th percentile of random kinematic encoders. Kinematic, language, and random targets do not transfer (-0.17 to -0.03). Two results limit the claim: a hand-picked kinematic descriptor, the stroke's initial direction, matches PMv better (0.65), and a label-shuffled PMv control scores 0.41. We specify the downstream manipulation test, on a CALVIN pipeline that reproduces Moto's published results, before running it.

cs.RO · cs.LG