跳到正文
研踪PaperTrail

YOUR RESEARCH, IN SIGHT

让下一篇重要论文，不再错过。

覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+2已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-08 10:17 · 全站最近入库：2026-11-08 06:02各信源进度 →

统计口径与数据新鲜度

统计 2026-11-07 13:25 — 2026-11-08 13:25（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：2026-11-08 10:17最近发现新增版本：2026-11-08 06:02库内论文最新更新时间：2026-11-06 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。

17,100 已收录论文14,661 AI 相关4,057 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

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 17,100 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-08 10:15（北京时间）
新论文 · v1arXiv:2611.03209

DynSL: Resource-Adaptive Model Training across Heterogeneous Edge Devices

Boris Radovič, Istemi Ekin Akkus, Marco Canini, Ruichuan Chen, Alice Dethise, Ivica Rimac, Veljko Pejović

原文摘要

Training models over data distributed across edge devices presents a trade-off: centralizing data simplifies training but raises privacy and regulatory concerns, while full local training may exceed the devices' compute and memory capacity. Split learning addresses this challenge by partitioning the model between clients and the server, allowing part of the computation to be offloaded. However, existing approaches typically rely on static split configurations or simplified assumptions that fail to capture the heterogeneity and complex dynamics of real deployments. We present DynSL, a framework that dynamically allocates model layers between clients and the server by continuously monitoring system metrics and periodically solving an optimization problem that accounts for client and server computation and communication. When beneficial, the system reconfigures the split, enabling the offloading strategy to adapt to changing device capabilities and system conditions. Extensive experiments on a testbed of 24 heterogeneous edge devices show that DynSL reduces training time by up to $\sim\!$30% over the strongest adaptive baseline and by more than 50% over a suboptimal uniform split, with no loss -- and sometimes a gain -- in accuracy.

cs.DC · cs.LG · cs.NI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03205

Optimal Sample Complexity for Max-Cut

Itamar Bender, Robert Krauthgamer

原文摘要

We determine the optimal sample size for constructing value coresets for Max-Cut via uniform vertex sampling. That is, sampling $O(\varepsilon^{-2})$ vertices uniformly at random in $n$-vertex graphs with edge weights in $[0,1]$ yields a reweighted subgraph whose Max-Cut value is an additive $\varepsilon n^2$-approximation to that of the original graph, with constant success probability. This improves the previously known $O(\varepsilon^{-4})$ upper bound and matches the known lower bound. The same guarantee holds also for importance sampling, where vertices are sampled proportionally to their weighted degrees, and we further prove a matching lower bound for this sampling method. Our analysis extends directly to metric Max-Cut, showing that importance sampling of $O(\varepsilon^{-2})$ vertices yields $(1\pm\varepsilon)$-approximation to the Max-Cut value, and to Max-Cut in graphs with edge weights in $[0,1]$ and average weighted degree $\bar d$, achieving the same $(1\pm\varepsilon)$-approximation by either uniform or importance sampling of $O(\varepsilon^{-2}n/\bar d)$ vertices. We additionally show how to use this coreset to recover a cut of expected value at least $OPT-\varepsilon n^2$, by making $O(n\varepsilon^{-2})$ additional edge-weight queries. Finally, our result extends to Max-$r$-Cut, giving an additive $\varepsilon n^2$-approximation using a sample of $O(\varepsilon^{-2}\log(2r))$ vertices.

cs.DS

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2512.05511

Rethinking Infrared Small Target Detection: A Foundation-Driven Efficient Paradigm

Chuang Yu, Jinmiao Zhao, Yunpeng Liu, Yaokun Li, Xiujun Shu, Yuanhao Feng, Bo Wang, Xiangyu Yue

原文摘要

While large-scale visual foundation models (VFMs) exhibit strong generalization across diverse visual domains, their potential for single-frame infrared small target (SIRST) detection remains largely unexplored. To fill this gap, we first systematically introduce frozen VFM representations into the SIRST task and propose a Foundation-Driven Efficient Paradigm (FDEP), a general and efficient framework compatible with diverse VFMs and SIRST networks, which improves detection accuracy without additional VFM-related inference overhead. Specifically, a Semantic Alignment and Modulated Fusion (SAMF) module is designed to achieve dynamic alignment and deep fusion of the global semantic priors from VFMs with task-specific features. Meanwhile, to avoid the inference-time overhead introduced by VFMs, we propose a Collaborative Optimization-based Implicit Self-Distillation (CO-ISD) strategy, which enables implicit semantic transfer between the main and lightweight branches through parameter sharing and synchronized backpropagation. In addition, to unify the fragmented evaluation system, we construct a Holistic SIRST Evaluation (HSE) metric that performs multi-threshold integral evaluation at both pixel-level confidence and target-level robustness, providing a stable and comprehensive basis for fair model comparison. Extensive experiments demonstrate that the SIRST detection networks equipped with our FDEP framework achieve state-of-the-art (SOTA) performance on multiple public datasets. Our code is available at https://github.com/YuChuang1205/FDEP-Framework

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2412.11161

Why and How: Knowledge-Guided Learning for Cross-Spectral Image Patch Matching

Chuang Yu, Yunpeng Liu, Jinmiao Zhao, Ao Chen, Xiujun Shu, Bo Wang, Zelin Shi, Xiangyu Yue

原文摘要

Recently, cross-spectral image patch matching based on feature relation learning has attracted extensive attention. However, existing methods focus on mining richer feature relations by building complex relation extraction structures. Meanwhile, performance bottlenecks have gradually emerged. To address this, we make the first attempt to explore a stable and efficient bridge between descriptor learning and metric learning, and construct a Knowledge-Guided Learning Network (KGL-Net), which achieves significant performance improvements while abandoning complex network structures. Specifically, we find that there is feature extraction consistency between metric learning based on feature difference learning and descriptor learning based on Euclidean distance. This provides the foundation for bridge building. To ensure the stability and efficiency of the constructed bridge, on the one hand, we conduct an in-depth exploration of 20 combined network architectures. On the other hand, a feature-guided loss is constructed to achieve mutual guidance of features. In addition, unlike existing methods, we consider that the feature mapping ability of the metric branch should receive more attention. Therefore, a hard negative sample mining for metric learning (HNSM-M) strategy is constructed. To the best of our knowledge, this is the first time that hard negative sample mining for metric networks has been implemented and brings significant performance gains. Extensive experimental results show that our KGL-Net achieves SOTA performance in multiple cross-spectral image patch matching datasets. Our code is available at https://github.com/YuChuang1205/KGL-Net

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03201

On the Convergence of Vanilla Adam under Heavy-Tailed Noise and Index Sampling

Kanata Oowada, Hideaki Iiduka

原文摘要

Understanding how stochastic gradient noise affects the convergence of Adam is an important problem in stochastic optimization, particularly when the noise exhibits heavy-tailed behavior. In this paper, we investigate the convergence of vanilla Adam in both the expected risk and the empirical risk minimization, focusing on the effects of heavy-tailed noise and its relationship to index sampling. For the expected risk minimization, we establish convergence guarantees under Hölder smoothness and bounded $\mathfrak{p}$-th moment assumptions, allowing stochastic gradient noise with infinite variance. Our analysis provides explicit upper bounds on the expected squared gradient norm and shows that the iteration-dependent convergence term decays more slowly as $\mathfrak{p}$ decreases. For the empirical risk minimization, we establish asymptotic and non-asymptotic convergence guarantees and demonstrate that heavy-tailed population noise can deteriorate convergence bounds through dataset-dependent constants as the dataset size increases. Furthermore, under Lipschitz smoothness, we formulate multistage Adam with practical learning rate schedules and demonstrate that increasing batch sizes can reduce index-sampling noise and improve convergence guarantees and SFO complexity. Numerical experiments support our theoretical findings on the effects of heavy-tailed noise and the computational benefits of increasing batch sizes.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03200

CISA: Retrofitting Dense Attention Models with Chunk-Indexed Sparse Attention

Mohammed Dabbah, Izhak Golan, Oren Tropp, Daniel Afrimi, Vladimir Anisimov, Tomer Asida, Nave Assaf, Ran El-Yaniv 等 13 位作者

原文摘要

Long-context LLM inference is costly because attention repeatedly reads a growing key-value (KV) history. We introduce Chunk-Indexed Sparse Attention (CISA), which retrofits pretrained dense attention models without gradient updates. CISA selects fixed-size KV chunks using compact summaries of existing keys, avoiding token-level indexing and a separate full-length index cache. It uses one indexing query per KV group and shares selected chunks across queries in that group. During prefill, QShare reuses selections across adjacent queries to reduce indexing cost and improve KV reuse and GPU utilization. We evaluate CISA on NVIDIA Nemotron 3 Nano and Ultra, and Inkling-Small. In one-layer BF16 benchmarks at batch size 32 using Ultra's per-device attention dimensions, core attention is 11.7x faster for prefill queries near the end of a 1M-token prompt and 13.0x faster during decode. Including chunk scoring, selection, and all other measured GPU work, speedups are 8.6x and 3.0x. On Ultra NVFP4 with a 4,096-chunk budget, aggregate scores stay within 1.5 percentage points of dense attention on MRCR, LongBench v2, and GraphWalks. RULER sweeps on Nano and Ultra BF16 up to one million tokens show scores within 2.05 percentage points of dense attention with 8-16x fewer KV reads. Serving million-token prompts with 32 concurrent requests and 4K output tokens per request, CISA achieves 2.29x the end-to-end throughput of dense attention using Nemotron 3 Ultra NVFP4 on four NVIDIA B300 GPUs. Selecting fewer chunks increases throughput to 2.61x; in separate GraphWalks evaluations, the smaller budget widens the aggregate score gap from 1.49 to 3.59 percentage points. These results expose a practical quality-speed tradeoff without sparse training.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03199

Cold-Start Pricing of LLM Services under Two-Sided Learning

Zhicheng Du, Pinyan Lu, Keyu Wang, Zihe Wang, Xiang Yan, Chunxue Yang

原文摘要

Online large language model (LLM) services face a pricing cold start: providers initially lack the historical data and user information needed for precise pricing, while users' willingness to pay evolves with repeated use. We formulate this challenge as a \emph{two-sided learning} problem between a provider and a user. The user learns from private usage experience, whereas the provider incurs a serving cost and dynamically adapts prices using only purchase decisions to maximize cumulative profit. The provider's central challenge is to learn to price against an evolving willingness to pay without observing the private experiences driving that evolution. We establish matching upper and lower bounds on the minimax regret over horizon $T$, identifying two cost-dependent regimes. When serving costs are sufficiently low, our Calibration-and-Tracking (CAT) algorithm achieves the optimal rate $Θ(T^{1/3})$. In the higher-cost regime, where the provider must learn whether continued service is profitable, Calibration-and-Tracking with Retirement (CAT-R) attains the optimal rate $Θ(\sqrt{T})$. This phase transition shows that serving costs affect not only profit margins but also the fundamental difficulty of learning to price LLM services. Simulations support the predicted regret scaling and show that our algorithms outperform standard online learning baselines in the evaluated settings.

cs.GT

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03198

Written Once, Read Differently How Language Models Own Answers Steer Their Next Ones

Marcus Armstrong

原文摘要

When a language model answers a yes/no question, that answer changes how it answers a different question about the same entity. We study this \emph{self-commitment} in 22 open-weight models from four families (0.5B--72B parameters), fixing the model's first answer and measuring the shift in its second across 3,600 question pairs about 120 entities. Each answer writes a \emph{commitment direction} into the residual stream. The similarity of two questions' directions predicts their coupling on held-out entities ($ρ= 0.43$--$0.85$), and ablating a single direction shared across topics leaves at most 17\% of the effect in every model. How this direction is read is far less stable: the reading circuit spreads across more attention heads as models grow, instruction tuning makes more questions read it in the opposite sense, and in one model changing only the system prompt reverses the circuit's sign. Ablating the direction while sparing the attention sink removes most of the coupling at under 1.3\% perplexity cost in 18 of 19 instruction-tuned models, though it shifts standalone answers in two. Without the sink spared, the direction's alignment with the leading principal component of ordinary text predicts which models pay a high cost. Written once, the commitment is read differently across scale, training and prompt.

cs.CL · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2403.11751

Relational Representation Learning Network for Cross-Spectral Image Patch Matching

Chuang Yu, Yunpeng Liu, Jinmiao Zhao, Dou Quan, Zelin Shi, Xiangyu Yue

原文摘要

Recently, feature relation learning has drawn widespread attention in cross-spectral image patch matching. However, existing related research focuses on capturing diverse feature relations between image patches and ignores sufficient intrinsic feature representations of individual image patches. To address this limitation, we propose an innovative relational representation learning that simultaneously focuses on sufficiently mining the intrinsic features of individual image patches and the feature relations between image patches. Based on this, we construct a Relational Representation Learning Network (RRL-Net). Specifically, we innovatively construct an autoencoder to effectively characterize the individual intrinsic features, and introduce a feature interaction learning (FIL) module to extract deep-level feature relations. Meanwhile, to further fully mine individual intrinsic features, a lightweight multi-dimensional global-to-local attention (MGLA) module is constructed to enhance the global feature extraction of individual image patches and capture local dependencies within global features. By combining the MGLA module, we further explore the feature extraction network and construct an attention-based lightweight feature extraction (ALFE) network. Furthermore, a multi-loss post-pruning (MLPP) optimization strategy is proposed, which can greatly facilitate network optimization while avoiding increases in parameters and inference time. Extensive experiments demonstrate that our RRL-Net achieves state-of-the-art (SOTA) performance on multiple public datasets. Our code is available at https://github.com/YuChuang1205/RRL-Net

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03197

Finding the Pivots to Repair Broken Source-Test Evolution Using LLMs

Xiao Chen, Songqiang Chen, Jialun Cao, Yun Peng, Chun Yong Chong, Shing-Chi Cheung

原文摘要

Repairing obsolete tests in response to source code evolution is critical for test suite reliability and software quality assurance. While large language models (LLMs) have emerged as promising tools for automating many software development tasks, the effective supply of context and the orchestration of repair pipelines in this source--test co-evolution scenario remain poorly understood. Moreover, although reasoning is found helpful for LLMs on some complex coding tasks, it remains unclear which reasoning steps actually contribute to effective test repairs and how the insights learned from LLMs' behaviors benefit coding agents. In this paper, we present a systematic empirical study of test repair for source-test co-evolution empowered by LLMs, investigating the effective setups and typical LLM behaviors. To support this study, we construct a new benchmark of 500 real-world source-test co-evolution cases from commits submitted between May 2024 and April 2026 across 15 highly-starred open-source Java projects. Our experimental results reveal several key insights for effective LLM-based test repair: the complementary benefits of combining test failure symptoms, source code changes, and supportive code context, as well as the effectiveness of self-repair rounds. We further derive a common reasoning action taxonomy from LLMs' reasoning traces, which proves instrumental in guiding LLMs to repair tests more effectively while at a lower financial cost. Applying our findings to the Codex agent further improves its repair success rate on challenging cases by 66.7%.

cs.SE

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03196

Dhārā Bench: An Evidence-Oriented Retrieval Benchmark for Indian Knowledge Systems

Avinash Holenarasipur Nagendra, Anand Kaniyaar Sriranga Prasad, Kousik Sarathy Sridharan, Mohan Raghavan

原文摘要

Dhārā Bench is a retrieval benchmark for Indian Knowledge Systems (IKS), with 325 questions in 12 categories and 60,121 text chunks from 15 source groups. The categories were developed based on 4,500 questions submitted to the Dhārā application; the evaluation set combines corpus-aligned application queries with expert-contributed questions. The annotations comprise 7,578 question-chunk relevance links involving 6,847 distinct corpus chunks. For 34 composite and multi-hop questions, 401 conjugate groups specify alternative sets of passages that jointly support each question. Chunk identifiers and hierarchical references connect the judgments to their source texts. The paper describes corpus preparation, annotation, and validation, and reports a fixed BM25 baseline implemented with Pyserini with $k_1=0.9$ and $b=0.4$. Across all questions, BM25 achieves MRR@20 of 0.315, Recall@10 of 0.182, and nDCG@10 of 0.205. On the group-annotated subset, complete-evidence scores are CE_Recall@20 = 0.294 and CE_MRR@20 = 0.141. The category results expose particularly weak verse retrieval, while group scores reveal gaps between finding individual passages and completing their support. The corpus and evaluation code are available at https://github.com/Avinash2236/Dhara-Bench.

cs.IR · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.15334

Concept-Grounded Reasoning with Prompt-Driven Localization for Interpretable Structured Report Generation

Xinyue Xu, Hongbin Lin, Juangui Xu, Hualiang Wang, Lehan Wang, Lijie Hu, Adrian Weller, Xiaomeng Li

原文摘要

Medical imaging modalities such as ultrasound and X-ray are widely used in clinical practice, where diagnosis follows a structured, evidence-driven workflow aligned with standardized criteria. While multimodal large language models (MLLMs) show promise for automated medical report generation, most existing systems rely on end-to-end multimodal fusion without modeling clinically defined intermediate attributes, leading to limited grounding and interpretability. To address this issue, we propose CORAL (COncept-grounded ReAsoning with Localization), a multimodal framework that integrates spatial grounding and concept-level supervision into a unified reasoning process. CORAL employs a prompt-driven medical segmentation model to localize lesions and predicts multi-class clinical attributes through a Concept Bottleneck module. The resulting textual concept tokens are combined with mask-modulated visual features within an MLLM to enable structured report generation and diagnostic prediction. Experiments on BUS-CoT and IU X-ray datasets demonstrate consistent improvements in diagnostic accuracy, concept consistency, and report quality over strong general-purpose and medical MLLMs, indicating that concept-grounded reasoning better aligns generation with clinical decision processes.

cs.CV · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2606.05506

Robust Scene Transfer for PointGoal Navigation via Privileged Sensor Guided Contrastive Learning

Amirhossein Zhalehmehrabi, Tiziano Tezze, Alberto Castellini, Alessandro Farinelli

原文摘要

We propose a sensor-guided adaptive contrastive learning framework for visual representation learning in PointGoal navigation. During training, privileged LiDAR sensing guides the contrastive objective through a geometry-aware similarity metric and adaptive temperature scaling, encouraging visual embeddings to capture navigation-relevant structure rather than scene-specific appearance. The resulting encoder is pretrained independently, frozen, and used as the perceptual backbone for reinforcement learning, decoupling representation learning from policy optimization. We further introduce a cross-stage domain mismatch between representation pretraining and policy learning to suppress environment-specific shortcuts and promote reliance on task-relevant features. Extensive experiments in high-fidelity simulation demonstrate that our approach significantly improves policy-level scene transfer across diverse indoor and outdoor environments. At deployment, the agent relies only on monocular RGB observations together with standard task-related inputs such as goal position and proprioceptive signals, without access to LiDAR or other privileged sensors. Our method outperforms large pretrained vision models and standard contrastive baselines under severe appearance and semantic shifts. We also release a multimodal dataset to support future research on privileged-guided visual representation learning for navigation. The code is available at:

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03194

Toward Deployable Hardware Roots of Trust for Edge AI: Software PUFs and Their Trade-offs

Yilin Li, Farhad Merchant

原文摘要

Deploying AI at the edge increases the need for security features on existing devices. When physical access to the device is available, storing secret keys in non-volatile memory is vulnerable. Physical unclonable functions (PUFs) address this by exploiting the unique yet repeatable characteristics of each device, but conventional implementations require design-stage modifications, making them inapplicable to a vast number of existing devices already manufactured. Recently, software PUFs have emerged as an alternative approach, leveraging analog characteristics exposed through software-only interfaces to derive unique responses, without post-fabrication changes. Since edge AI hardware is bound to long operational lifespans, this offers clear economic and engineering value. Existing research on software PUFs remains fragmented. It lacks a holistic review of the trade-offs among choices of components used as entropy sources and software methods to expose analog characteristics. This gap hinders assessing feasibility, comparing approaches, and identifying open challenges towards a hardware root of trust in existing devices. This paper surveys 37 existing software PUF works and compares them against the requirements of on-device model protection: boot-time versus runtime key derivation, and response latency. It analyzes trade-offs, outlines four key challenges that current software PUFs face for edge AI deployment, and highlights promising directions for overcoming barriers to real-world adoption.

cs.CR

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.01549

NashDreamer: Model-Based Reinforcement Learning for Zero-Sum Imperfect-Information Games

Tomáš Holeček, Viliam Lisý

原文摘要

Model-based reinforcement learning (MBRL) has achieved remarkable results in single-agent domains, yet its extension to competitive imperfect information games (IIGs) remains underexplored. In multi-agent settings, opponent-induced non-stationarity complicates the learning process, and decentralized model learning faces severe identifiability barriers, which we argue make centralized model learning a mathematical necessity. Building on this analysis, we propose NashDreamer, a principled MBRL framework for two-player zero-sum IIGs. It introduces a centralized Multi-Agent Recurrent State-Space Model (MARSSM) that decouples environment dynamics from the effect of players' strategies on their individual observations. NashDreamer is designed to use arbitrary policy gradient algorithms and inherits their convergence guarantees towards Nash equilibria under an idealized model. Empirical evaluations across five benchmark games demonstrate that NashDreamer substantially improves sample efficiency over model-free baselines early in the training. Finally, we theoretically analyze the architecture's optimization landscape, identifying the vulnerability of the Dreamer family of algorithms to posterior collapse in stochastic environments. We leave it as an open challenge.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03193

Harnessing Mutual Coupling in Cell-Free mMIMO

Erind Tufa, Antonio A. D'Amico, Luca Sanguinetti

原文摘要

This letter investigates a cell-free massive MIMO system with $L$ access points (APs), each employing a small fixed-form-factor, uniform linear array of $N$ potentially closely spaced antennas. Utilizing multiport communication theory, we analyze the impact of mutual coupling on uplink and downlink spectral efficiency (SE) under various impedance-matching strategies. We compare two deployment regimes: ($i$) increasing $N$ while $L$ is kept fixed, and ($ii$) increasing $N$ while fixing the total antenna count $M = NL$. Numerical results show that, with a full matching network, mutual coupling itself becomes beneficial when an AP serves several user equipments, which is the typical operating regime of user-centric cell-free networks. Furthermore, with full matching, antenna densification can enhance SE or allow for a reduction in the number of APs without performance loss. Since full matching networks remain of manageable size in the small arrays of distributed APs, unlike in large centralized arrays, cell-free networks offer a natural setting for exploiting mutual coupling.

cs.IT · eess.SP

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03192

Verification Allocation in Stylized LLM Agent Pipelines under Hidden Tool Degradation: A Discrete-Time Mean-Field Analysis

Shereen Anand, Chandramani Singh

原文摘要

We study a stylized verification-allocation model for role-structured pipelines with binary inherited corruption and hidden tool degradation. Standard contraction arguments give serial and multi-role fixed points and a finite-population deviation bound under independence. When verification actions affect neither future errors nor observations, the belief-state choice is myopic. We characterize a two-action threshold under an explicit monotonicity condition and a restricted four-action interval and tail policy when generation rates are equal and only one check degrades. Synthetic experiments check these model-specific results and find counterexamples outside the restriction. This is conditional model analysis, not a general mean-field-game result or empirical evidence about deployed LLM systems.

cs.GT

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.25391

Rapid Mixing for Sampling Colorings of Random Graphs up to Tree Uniqueness Threshold

Zhidan Li, Kuan Yang

原文摘要

We study the mixing time of the single-site Glauber dynamics for sampling proper $q$-colorings on sparse Erdős-Rényi random graphs $\mathcal{G}(n, d/n)$. We prove that, whenever $q \ge (1 + o_d(1))d$, Glauber dynamics mixes in polynomial time with high probability over $\mathcal{G}(n, d/n)$ for all sufficiently large integers $n$. This improves the previous bound $q = (α^* + o_d(1))d$ where $α^* = 1.7632...$ is the unique solution of $x = \exp(1/x)$ on $[1, \infty)$ by Efthymiou, Hayes, Štefankovič, and Vigoda (SODA, 2018) and asymptotically matches the uniqueness threshold on infinite regular trees. Our proof has three main ingredients. First we establish logarithmic bounds on the spectral and coupling independence constants of the Gibbs distribution by extending the tree analysis of Chen, Liu, Mani, and Moitra (J. ACM, 2026). Second, we extend the entropy-factorization framework for two-spin systems on random graphs, developed by Bezáková, Galanis, Goldberg, and Štefankovič (TALG, 2024), to proper colorings by introducing a buffer around high-degree vertices. Finally, we compare the resulting heat-bath block dynamics with the single-site Glauber dynamics.

math.PR · cs.DM · cs.DS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03190

Asymptotically Optimal Weighted Edge Colorings via Local Bin Packing

Rishikesh Gajjala, Arindam Khan

原文摘要

We study weighted edge coloring, where each edge of a graph $G$ has a weight in $(0,1]$, and colors are assigned so that the incident weight of each color at every vertex is at most one. The objective is to minimize the number of colors, denoted by $χ'_w(G)$. This generalizes classical edge coloring (all weights are one) and bin packing ($G$ is a star). Two natural lower bounds are the maximum weighted degree $Δ_w(G)$ and the maximum local bin-packing number $\mathrm{bp}(G)$: the maximum, over vertices, of the minimum number of unit bins needed to pack the incident weights. We have $\lceilΔ_w(G)\rceil\le\mathrm{bp}(G)\leχ'_w(G)$. How closely these local lower bounds approximate the global optimum is a longstanding open problem. Our main result establishes an asymptotic weighted analogue of Vizing's theorem: every simple graph admits a coloring using at most $(1+o(1))\mathrm{bp}(G)$ colors. Thus local bin packings can be made globally consistent with asymptotically vanishing relative loss, even in nonbipartite graphs. Our local-to-global rounding framework coordinates feasible local packing distributions into a partial coloring, leaving only $o(\mathrm{bp}(G))$ uncolored weight at every vertex. For bipartite multigraphs, we prove $χ'_w(G)\le(2+o(1))Δ_w(G)$, establishing the Chung--Ross conjecture [SICOMP'91] asymptotically. For loopless multigraphs, we obtain the exact bound $χ'_w(G)\le3\lceilΔ_w(G)\rceil$. Both leading coefficients are optimal, giving weighted analogues of the classical Kőnig--Shannon bounds. Further upper and lower bounds show that the classical factor-$3/2$ separation does not persist with the local bin-packing benchmark. We also disprove the exact weighted-degree conjecture of Feige and Singh [ESA'08]. Our upper bounds have polynomial-time algorithms, and their existence statements are formalized in Lean 4.

math.CO · cs.DM · cs.DS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03189

MoSoLoco: Monocular 3D Human Localization through Visual Latent Features and Social Context

Yasamin Borhani, Javad Khoramdel, Yang Gao, Po-Chien Luan, Alexandre Alahi

原文摘要

Estimating the metric 3D position of people from a monocular camera is important for robot navigation and motion prediction. Lightweight pose-based localizers are attractive for robotic hardware with limited computational resources because they operate on compact 2D keypoints. However, they typically discard the richer visual representations computed by the pose estimator. We introduce MoSoLoco, a per-frame monocular 3D human localizer that combines the pose detector's internal visual features with the detected 2D skeleton, without requiring an additional RGB or depth encoder. MoSoLoco can further incorporate sparse same-frame multi-person context to refine individual position estimates. Experiments on JRDB and NuScenes show that reusing pose-detector representations improves localization, reducing the best JRDB error by approximately 23 percent relative to the strongest evaluated keypoint-based baseline. A sparse social-interaction module incorporates nearby detections from the same image for further refinement. Controlled trajectory-forecasting experiments show downstream benefits, while the complete pipeline remains suitable for efficient Jetson AGX Xavier inference.

cs.CV · cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03186

The Covering Radius of the Reed--Muller Code \texorpdfstring{$\RM(3,8)$}{RM(3,8)} is 56

Zhihang Zhao, Yansheng Wu

原文摘要

We prove that the covering radius of the third-order Reed--Muller code in eight variables is fifty-six. The proof separates even- and odd-weight inputs by moment-controlled slicing. On the even side, a transport invariant on quadratic-label stars and consistency of singleton restrictions give a compatible cubic correction. On the odd side, symmetry of the third moment yields a covariance-rank obstruction. Its rank-two exception is a quartic plus three points of a two-plane; the zero-second-moment case is a prescribed-point perturbation of a quintic. We prove that every such point perturbation has third-order nonlinearity at most fifty-five by a marked derivative identity and complete restriction-code covers. Consequently the maximum among exact-degree-eight functions is fifty-five. The structural reductions are proved analytically, and complete finite certificates verify the explicitly stated local covering and compatibility assertions. The matching lower bound follows from the published quartic relative radius.

cs.IT

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03184

ACTOR: Action-Centered Temporal Organization for Literary Role-Playing Agents

Yuxiang Gao, Tianjie Ju, Zheng Wu, Zongru Wu, Hao Fei

原文摘要

Large language agents for literary role-playing are typically modeled around dialogue generation, leaving characters' non-dialogue behaviors and their dynamic evolution over story time underrepresented. In this paper, we propose ACTOR (Action-Centered Temporal Organization for Literary Role-Playing Agents), an automated framework for constructing such behavior trajectories from raw novel texts. ACTOR extracts dialogue and persona-relevant non-dialogue behaviors, represents each behavior using a Thought-Action-Observation structure, and organizes behavior trajectories according to the internal temporal order of the story. We build a behavior dataset from 143 diverse literary works. For each source behavior, we generate counterfactual behaviors that are contextually plausible yet inconsistent with the character's persona. We find that existing models struggle to reliably infer character-consistent actions from historical behaviors and evolving character states. Using these behavior pairs, we train models with Regularized Preference Optimization (RPO), which augments DPO with a negative log-likelihood loss on the source behavior to jointly optimize behavior preference and generation. We further design a four-way next-behavior selection task to evaluate the model's ability to faithfully mimic character actions. Across 17,560 evaluation instances, the model trained on ACTOR's preference pairs achieves substantially improved accuracy. This result demonstrates that structured, temporally organized supervision enables models to leverage character profiles and prior behaviors to infer subsequent actions, greatly advancing literary role-playing beyond mere dialogue imitation. Code and data are available at https://github.com/SilentStar233/ACTOR.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03181

Multi-GPU Acceleration of Interface Problems in FEniCS, with Application to Cardiac Modeling

Edoardo Centofanti, Benjamin Pachev, Joshua Steyer, James D. Trotter, Pietro Benedusi

原文摘要

GPU acceleration of finite element assembly is well established for standard formulations, but mixed-dimensional problems, in which volumetric subdomains are coupled via lower-dimensional interfaces, remain unsupported in automated frameworks such as FEniCS. We extend cuDOLFINx, a GPU-enabled extension of the DOLFINx finite element framework, to assemble such interface problems on GPUs while preserving the high-level FEniCS programming model. Combined with GPU-resident Krylov solvers and algebraic multigrid preconditioning, this enables end-to-end GPU execution of the EMI (Extracellular--Membrane--Intracellular) model, an elliptic interface problem from cardiac modeling in which many intracellular subdomains are coupled through membranes. We compare the GPU implementation against a CPU-based one through strong and weak scaling experiments on regular and realistic 3D cardiac geometries with up to 8000 cells and 22 million degrees of freedom. In terms of assembly, a single GH200 node is roughly equivalent to two 144-core CPU nodes in strong scaling, and to two to three nodes in weak scaling, depending on geometry. For the linear solver, performance depends on geometry, preconditioning, and problem scale. On the regular geometry, GPU advantages emerge mainly at larger problem sizes and node counts. On the realistic geometry, with the same algebraic multigrid preconditioner for both architectures, the GPU is faster at all tested scales. The implementation is available as open-source software and applies to other problems with interface coupling.

math.NA · cs.DC · cs.MS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03179

Every Clause Counts: Instruction-Following Reranking by Multiplying Small, Answerable Questions

S. M. Mahbub Uddin Ahmed, Abdullah Al Musaddiq

原文摘要

People often attach instructions to their search queries, asking for pages written for beginners, documents in a certain language or official sources only. Most rerankers were trained to judge topical relevance and treat such instructions as a few extra words. On the InfoSearch benchmark, even GPT-4o used as a list-wise reranker reaches an average WISE (Weighted Instruction Sensitivity Evaluation) of only 33.5. We present DMReranker (Decision-Model Reranker), a zero-shot reranker that splits an instructed query into a topic and a short list of typed clauses. A small judge model answers one narrow question per clause, and the answer probabilities are multiplied into a single score. Nothing is trained, and the configuration was frozen before the test runs. Reranking the same top-100 candidates as the published baselines, DMReranker reaches a macro WISE of 50.9 over all six InfoSearch dimensions while keeping original-mode nDCG@10 at 82.9. It trails GPT-4o on the stricter SICR (Strict Instruction Compliance Ratio) metric, 19.2 against 32.1. We show that this gap follows from how a product of probabilities behaves when SICR compares raw scores across runs. Computed from ranks, our SICR is 67.0. On the same judge, splitting the instruction raises WISE from 32.9 to 47.6 over a single combined question, while raw-score SICR falls. Scoring all six dimensions cost about $10 at list price.

cs.IR · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03178

CoworkOS:Scaling Agent Coworkers for Real-world Automation

Jiajie Jin, Guanting Dong, Chengjun Pan, Xin Deng, Zhuohui Sheng, Guoqiang Wang, Tao Gui, Yutao Zhu 等 40 位作者

原文摘要

Rapid advances in cowork agents are shifting human-AI collaboration toward real-world automation, delivering software, designs, and other artifacts. However, such productivity work demands both high-quality delivery on complex productivity tasks and continued monitoring and decision-making over long horizons, a combination that remains an open challenge. We introduce CoworkOS, a general-purpose harness for long-horizon work that separates task strategy from execution management: the model decides actions, delegation, and acceptance, while the runtime maintains state and feedback. Its three-layer architecture enables direct execution or delegation to Subagents, synchronously or asynchronously, under stable responsibilities. Through a shared application layer, humans and agents inspect and revise the same evolving documents, presentations, CAD models, and Blender scenes, so human-in-the-loop participation can enter both through conversation and through direct modification of the work itself. To sustain continuity across agents and sessions, CoworkOS further introduces hierarchical workspaces and persistent semantic memory, so working files, key decisions, and long-running Job logs remain shareable, reusable, and resumable after interruption. Across 12 benchmarks and 6 categories of real-world applications, CoworkOS proves highly competitive with frontier agent applications such as Codex and DeepSeek Harness. As a sustained-work case study, CoworkOS uses GLM-5.3 as its backbone to conduct RSI-driven automated post-training on Qwen3.6-35B-A3B. Over five rounds, the model's average score across eight benchmarks rose from 33.19 to 42.95 (+9.76). Looking ahead, we believe CoworkOS opens a meaningful path toward general cowork-agent intelligence.

cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03177

Functional Logic Programming: Improving Logic Programs with Ease

Michael Hanus

原文摘要

Logic programming is based on defining relations. Functions are often considered as syntactic sugar which can be transformed into predicates so that their logic is not used for computational purposes. In this paper, we show that the use of functions improves the operational behavior of logic programs without loosing the flexibility of logic programming. For this purpose, we present a systematic transformation from predicates and goals into functions and nested expressions, respectively. By evaluating functions in a demand-driven manner and taking potential failures into account, it is ensured that the execution of the transformed programs will never require more steps than the original programs but can decrease the number of steps---in the best case reducing infinite search spaces to finite ones. Thus, functional logic programming provides an adequate basis to improve the operational behavior of logic programs without changing their semantics. This also shows that, from an operational point of view, functional logic programming has clear advantages compared to relation-based logic programming.

cs.PL

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2502.04379

Can Large Language Models Capture Video Game Engagement?

David Melhart, Matthew Barthet, Georgios N. Yannakakis

原文摘要

Can out-of-the-box pretrained Large Language Models (LLMs) detect human affect successfully when observing a video? To address this question, for the first time, we evaluate comprehensively the capacity of popular LLMs for successfully predicting continuous affect annotations of videos when prompted by a sequence of text and video frames in a multimodal fashion. In this paper, we test LLMs' ability to correctly label changes of in-game engagement in 80 minutes of annotated videogame footage from 20 first-person shooter games of the GameVibe corpus. We run over 4,800 experiments to investigate the impact of LLM architecture, model size, input modality, prompting strategy, and ground truth processing method on engagement prediction. Our findings suggest that while LLMs rightfully claim human-like performance across multiple domains and able to outperform traditional machine learning baselines, they generally fall behind continuous experience annotations provided by humans. We examine some of the underlying causes for a fluctuating performance across games, highlight the cases where LLMs exceed expectations, and draw a roadmap for the further exploration of automated emotion labelling via LLMs.

cs.CV · cs.AI · cs.CL · cs.HC

查看论文
arXiv ↗PDF ↗
版本更新 · v5arXiv:2605.05155

Aes3D: Aesthetic Assessment in 3D Gaussian Splatting

Chuanzhi Xu, Boyu Wei, Haoxian Zhou, Xuanhua Yin, Zihan Deng, Haodong Chen, Qiang Qu, Weidong Cai

原文摘要

As 3D Gaussian Splatting (3DGS) gains attention in immersive media and digital content creation, assessing the aesthetics of 3D scenes becomes important in helping creators build more visually compelling 3D content. However, existing evaluation methods for 3D scenes primarily emphasize reconstruction fidelity and perceptual realism, largely overlooking higher-level aesthetic attributes such as composition, harmony, and visual appeal. This limitation comes from two key challenges: (1) the absence of general 3DGS datasets with aesthetic annotations, and (2) the intrinsic nature of 3DGS as a low-level primitive representation, which makes it difficult to capture high-level aesthetic features. To address these challenges, we propose Aes3D, the first systematic framework for assessing the aesthetics of 3D neural rendering scenes. Aes3D includes Aesthetic3D, the first dataset dedicated to 3D scene aesthetic assessment, built on our proposed annotation strategy for 3D scene aesthetics. In addition, we present Aes3DGSNet, a lightweight model that directly predicts scene-level aesthetic scores from 3DGS representations. Notably, our model operates solely on 3D Gaussian primitives, eliminating the need for rendering multi-view images and thus reducing computational cost and hardware requirements. Through aesthetics-supervised learning on multi-view 3DGS scene representations, Aes3DGSNet effectively captures high-level aesthetic cues and accurately regresses aesthetic scores. Experimental results demonstrate that our approach achieves strong performance while maintaining a lightweight design, establishing a new benchmark for 3D scene aesthetic assessment.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03174

Final coalgebras over R-modules

Oisín Flynn-Connolly, Henning Basold, Chase Ford

原文摘要

Final coalgebras and the coinduction principle for coalgebras over $R$-modules, where $R$ is a commutative ring, are essential in determining the denotational behaviour of various types of linear weighted automata. The goal of this paper is to provide lower and upper bounds on the ordinals that are needed to construct final coalgebras over $R$-modules as the limit of a transfinite sequence of functor iterations. We give an example of a binary intersection preserving finitary functor on the category of abelian groups, which is the category of modules over integers, that fails to converge in $ω\cdot 2$ steps, thus settling an open problem of Adámek, Milius, and Moss. Then we relate this to the failure of the Artinian property of the integers and show that all finitary functors on $R$-modules for a Noetherian ring $R$ with Krull dimension $n$, which preserve countable directed intersections and a generalised notion of monomorphism, converge in $ω\cdot (n + 2)$ steps. We also establish how the connection between Aczel-Mendler bisimulation and behavioural equivalence is mediated by the functor $\operatorname{Ext}^1$. These results show that homological algebra tools can play an important role in coalgebra.

cs.LO · math.AC · math.CT

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03173

PagedMoE: Cross-Precision Expert Caching for Token-Level Dynamic Precision in SSD-Resident MoE Inference

Yutong Lin, Yingdi Shan, Shaoyuan Chen, Yuanyong Chen, Yongwei Wu, Mingxing Zhang

原文摘要

Large Mixture-of-Experts (MoE) models are promising for local inference because sparse routing allows the routed-expert pool to reside on SSD while DRAM caches recently used experts. However, under practical DRAM cache capacities, substantial SSD I/O remains on the critical path despite I/O-computation overlap. Token-level dynamic expert precision offers a natural way to reduce this data demand by allocating bits according to expert importance and quantization sensitivity. Nevertheless, treating each bit-width of the same expert as an independent object prevents cross-precision cache reuse, causing overlapping data to be fetched repeatedly. The key insight is that precision should instead denote the refinement depth of one shared expert state. Based on this insight, we present PagedMoE, which stores each expert once using a prefix-compatible tree codebook and bitplane layout. Lower-precision representations are valid prefixes of higher-precision ones, allowing cached data to be reused or extended by fetching only missing bitplanes. PagedMoE selects the required prefix length before I/O using current routing weights and offline expert-specific error profiles, and directly executes cached 1-8-bit prefixes without materializing full-precision experts. At an average routed-expert precision of 2.5 bits, PagedMoE achieves quality close to GPTQ-INT4 on Qwen3.5-397B-A17B and to MXFP4 on most evaluated GLM5.2 tasks. With a 24 GiB expert cache, it reaches 8.89 tokens/s on Qwen3.5, 1.33x the throughput of the same dynamic requests using independent precision objects and 1.56x that of uniform 4-bit execution.

cs.AR

查看论文
arXiv ↗PDF ↗