cs.AI · 2026-11-05 · No. 164

Artificial Intelligence, 2026-11-05.

46 new papers in cs.AI. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

46 entries
  1. 01

    BEAKER: An Expert-Curated Benchmark for Embodied Brains in Self-Driving Chemical Laboratories

    Fei Lin, Tengchao Zhang, Ziyang Gong, Xiaotong Yu, Bohan Zhang, Yifan Zhou, Qihao Yang, Ji Dai, Dong Li, Yining...

    cs.AI

    Self-Driving Chemical Laboratories (SDCLs) are moving chemical experimentation toward embodied automation, where machines must perceive laboratory scenes, reason about experimental states, and act under physical, procedural, and safety constraints. Although Multimodal Large Language Models (MLLMs) are increasingly considered as embodied brains for such systems, existing laboratory benchmarks mainly focus on safety, anomaly detection, or...

    arxiv.org/abs/2611.06911 · PDF

  2. 02

    FrontierPhysics: Evaluating Agents for End-to-End Physics Research

    Bingran You, Wenjun Ke, Ria Rosenauer, Bo Liu, Wenbo Chen, Min Chen, Shuqi Xu, Yue Ma, Jiechao Feng, Bo Chen,...

    cs.AI

    AI agents are playing an increasingly important role in physics research. However, existing physics benchmarks only evaluate fixed questions, isolated sub-tasks, or paper reproduction with given solutions, and rarely check research quality alongside correctness. We introduce FrontierPhysics, an end-to-end benchmark with 56 tasks made by 42 physicists based on their own research, covering nine domains across theoretical, computational, and...

    arxiv.org/abs/2611.06910 · PDF

  3. 03

    Dull, Dirty, Dangerous? Working Conditions and the Limits of Occupational AI Exposure

    Siyi Yu

    cs.AI · cs.HC · econ.GN

    Automation is often justified as relief from dull, dirty, and dangerous work. I examine how these conditions relate to occupational AI exposure using 21 O*NET items across 754 US occupations. Higher exposure is associated with lower condition scores under a large language model (LLM) measure and AI Occupational Exposure (AIOE), with equal occupation weights and employment weights. Adjustment for physical demands, required education, training...

    arxiv.org/abs/2611.06888 · PDF

  4. 04

    Learning to Decode Linear Codes with Generalized Belief Propagation Network

    Samira Malek

    cs.AI

    Error correcting codes are a fundamental component of communication systems, enabling reliable data transmission over noisy channels. Recently, neural decoders have demonstrated notable performance gains over classical iterative algorithms. In this paper, we propose a Generalized Belief Propagation Network (GBPNet) that preserves message-passing symmetry while enhancing the decoder's learning capability. The architecture uses cascaded...

    arxiv.org/abs/2611.06874 · PDF

  5. 05

    On Turn-Level Best-of-N Sampling for Scaling LLM Agents

    Jalaj Bhandari, Dhawal Gupta, Hongli Zhan, Zhenting Wang, Varad Pimpalkhute, Zhengzhong Liu, Eric Xing, Rupesh Srivastava

    cs.AI

    In this work, we evaluate a simple inference-time scaling approach based on \textit{turn-level best-of-N sampling} (\method) for large language model (LLM) agents on complex, long-horizon tasks. We instantiate this using a \textit{generator-selector} framework with two LLMs; at each turn, a \emph{generator} model generates $N$ candidate actions and a \emph{selector} model selects one of them, which is then executed in the environment. This...

    arxiv.org/abs/2611.06868 · PDF

  6. 06

    CompressionBench: How can your coding agents compress LLMs?

    Xinrui Zhong, Zhifeng Li, Zhiming Wang, Yongqi Liu, Shanglin Yang, Yang Zhou, Zikun Li, Zhuoming Chen, Beidi Chen

    cs.AI

    Recursive self-improvement requires AI systems to improve the models and infrastructure used by subsequent systems. LLM compression offers a concrete efficiency objective: retain a pretrained model's behavior within a smaller model storage budget. We study whether coding agents can carry out this step autonomously. We formulate compression as a terminal task in which an agent receives a source checkpoint, storage limit, inference environment,...

    arxiv.org/abs/2611.06858 · PDF

  7. 07

    ServerBench: How can your coding agents design specialized LLM serving systems?

    Xinrui Zhong, Zhiming Wang, Yongqi Liu, Zhaohui Wang, Shanglin Yang, Zikun Li, Zhuoming Chen, Beidi Chen

    cs.AI

    Large language model (LLM) coding agents develop software through experimentation, offering an opportunity to design the serving systems supporting their own operation. Effective serving requires coordinating implementation and operating decisions under session dependencies, latency constraints, and correctness requirements. We introduce ServerBench, a benchmark for agentic design of complete, specialized LLM serving systems. Agents...

    arxiv.org/abs/2611.06841 · PDF

  8. 08

    Rules, Traces and Routing Statistics: Training and Calibration Data for Open Models on ARC-AGI-2

    Nguyen Minh Thang, Trinh Thi Phuong

    cs.AI

    We study which training and calibration data help three open models solve ARC-AGI-2 tasks when the solver runs offline on four L4 GPUs for at most 12 hours. We compare data offline on public tasks and report 189 runs on the 120 hidden semi-private tasks. For the NVARC Qwen3-4B grid model, we built a corpus of 12,000 executable rules in a typed language and 95,993 tasks; a rule is admitted only if its behaviour on a fixed probe set differs...

    arxiv.org/abs/2611.06827 · PDF

  9. 09

    Task-Efficient Agent Evaluation Using Cached Traces

    Ben Johnson, Hayden Helm

    cs.AI

    Efficient benchmarking estimates a system's full benchmark performance from a small number of task executions. Standard approaches rely solely on the binary correctness of executions from a collection of already-evaluated reference systems. An agentic system, however, produces a rich trace for each execution that can reveal substantially more about its capabilities. Leveraging this information requires useful trace representations. We propose...

    arxiv.org/abs/2611.06803 · PDF

  10. 10

    ConceptARC at Scale: Studying Conceptual Understanding in Reasoning Models

    Philipp Mondorf, Adrien Deliège, Solim LeGris, Claas Beger, Ryan Yi, Barbara Plank, Brenden M. Lake, Todd M. Gureckis

    cs.AI · cs.LG

    Large reasoning models (LRMs) perform well on challenging benchmarks, yet it remains unclear whether they form generalizable abstractions across conceptually related reasoning tasks. The Abstraction and Reasoning Corpus (ARC) provides a controlled testbed for studying abstraction, requiring solvers to infer transformation rules over two-dimensional grids from a small set of examples. While ARC evaluates rule abstraction within individual...

    arxiv.org/abs/2611.06732 · PDF

  11. 11

    LifeVal: Evaluating AI Model Performance on Personalized Agentic Consumer Tasks

    Jenny Yang, Elizabeth Proehl, Anjan Bharadwaj, Emily Fan, Charlotte Cole, James Kim, Noam Brown, Phoebe Thacker,...

    cs.AI

    We introduce LifeVal, a benchmark evaluating AI model performance on personalized agentic consumer tasks. LifeVal covers the most valuable life tasks that a personalized digital superassistant could automate, representing over 1.7 trillion US dollars in annual willingness-to-pay across 148 consumer task categories (e.g., managing finances, moving apartments, searching for a job, finding childcare). To construct our activity distribution, we...

    arxiv.org/abs/2611.06687 · PDF

  12. 12

    Where Search Agents Go Off the Rails: Graph-Grounded Step-Level Verification

    Hyeseo Jeon, Seonho Yoo, Jay-Yoon Lee

    cs.AI

    Search-augmented reasoning systems can reach correct answers while making errors in intermediate steps. Reliable auditing must check whether evidence $D$ covers the question $Q$, reasoning $T$ addresses $Q$, and $T$ is grounded in $D$. We introduce GraphRail, which represents requirements, evidence, and reasoning as structured claims and compares them without a generative judge or gold answer. Across three multi-hop QA benchmarks and three...

    arxiv.org/abs/2611.06662 · PDF

  13. 13

    From Certified Pairs to a Certified Hub: Reasoner-Checked Crosswalks for the Information Exchange Standard

    Fabio Rovai

    cs.AI · cs.DB · cs.LO

    A national digital twin was conceived as an ecosystem of twins connected through mappings to a common foundation, but the roadmap that promises to confirm those mappings are valid never says what valid means. We give the test a precise form, coherence and conservativity of the merged ontology, borrowed from logic-based alignment repair, and apply it to the Information Exchange Standard (IES), first pair by pair and then to whole hubs....

    arxiv.org/abs/2611.06628 · PDF

  14. 14

    Interactive World Simulation with Cosmos for Physical AI

    NVIDIA, :, Anis Ahmad, Jon Allen, Malik Aqeel Anwar, Jesse Archer, Maciej Bala, Piotr Barejko, Aarti Basant, Dan...

    cs.AI · cs.CV · cs.LG · cs.RO

    We present a shared framework that adapts Cosmos world foundation models for real-time interactive world simulation in Physical AI. By streaming visual observations in response to actions and scene controls, the framework enables users and embodied agents to interact with evolving virtual environments. It supports both predicting the visual consequences of actions and generating observations from physical state maintained by an external...

    arxiv.org/abs/2611.06627 · PDF

  15. 15

    Time-Division Multiplexing: The Operational Rhythm and Temporal Organization Layer of Digital Life

    Haibin Tong, Luming Guo, Nana Cui, Jiang Yu

    cs.AI · q-bio.NC

    Continual agents must handle urgent interaction while retaining alternatives, replaying errors, and committing safe long-term changes. A previous digital-life framework distinguishes online (busy) and offline (idle) channels, but leaves open whether these are separate programs or phases of a common process and how their operations should be ordered. Motivated by recent observations of rapidly alternating perceptual representations in primate...

    arxiv.org/abs/2611.06603 · PDF

  16. 16

    One Scale, Four Texts: Gaps, Overlaps and Silent Repairs in Published Grading Schemes

    Elisa Rovai, Fabio Rovai

    cs.AI

    Category systems often reach their users as text, and readers apply them with tacit repairs. We ask what such a text determines once the repairs are removed, using diabetic retinopathy grading, where one international scale circulates in several documents. We transcribe four texts (the English screening definitions of 2025, the international scale as tabulated in 2017 and as published in 2003, and an Italian guideline of 2015) into one rule...

    arxiv.org/abs/2611.06569 · PDF

  17. 17

    RAGAM-AI: A Hybrid Reinforcement Learning based Agentic approach powered by Generative AI for Edge Resource Management

    Shreya K. Chari, John S. Vardakas, Kostas Ramantas, Christos Verikoukis

    cs.AI · cs.LG

    Resource management for distributed edge networks hosting different services has been handled efficiently using Reinforcement learning (RL). Reward generalization in complex workflows is still an open issue within the RL realm. Therefore, this work proposes RAGAM-AI, a hybrid framework that combines the capabilities of RL algorithms and Generative-AI (Gen-AI) for powering an Agentic AI workflow that optimizes resource allocation at the...

    arxiv.org/abs/2611.06554 · PDF

  18. 18

    Addition by Subtraction: Toward Mitigating Fragile Safety Alignment in Large Language Models

    Hamid Dashtbani, Alexandra Dmitrienko

    cs.AI · cs.CR

    Large language models (LLMs) undergo safety alignment to prevent harmful generations, yet they remain vulnerable to jailbreak attacks. Recent work reveals a deeper structural weakness: safety alignment can be highly sparse, such that pruning a small number of safety-critical neurons or ablating a single refusal direction can substantially undermine aligned behavior. In this work, we take the first steps toward strengthening safety alignment...

    arxiv.org/abs/2611.06553 · PDF

  19. 19

    Using Experience Twice: Dynamic Process Rubrics for Privileged Token-Level Credit Assignment

    Shumao Wu, Kongcheng Zhang, Jianqiang Huang, Baisheng Lai

    cs.AI

    Agent trajectories often contain both successful decisions and failed attempts, yet optimization with sparse trajectory-level rewards struggles to identify which critical decisions should be reinforced or suppressed. We observe that meaningful distinctions among on-policy trajectories can be distilled into natural-language process rubrics and used as process supervision signals. Based on this insight, we propose Online Rubric...

    arxiv.org/abs/2611.06543 · PDF

  20. 20

    ElecLoad: Adaptive Electricity Load Forecasting Guided by Multiple Events

    Huajian Zhang, Mingyue Cheng, Xingpeng Gao, Xiaoyu Tao, Zhiding Liu, Jiqian Yang, Zhi Li

    cs.AI · cs.LG

    Accurate electricity load forecasting provides critical decision support for energy procurement, generation scheduling, and supply-demand balancing, contributing to the economic efficiency and operational reliability of power systems. However, future electricity demand can be affected by the complex interplay of multiple events described in Web notices, weather alerts, and event schedules. Their effects vary over time and can reinforce or...

    arxiv.org/abs/2611.06508 · PDF

  21. 21

    Tools Need Not Speak in Text: Latent Context Compression for Agents

    Yuetong Liu, Zhehao Zhou, Yitong Zhang, Haorui Chen, Jia Li

    cs.AI

    Large language model (LLM) agents typically append raw tool outputs to their context. Over multi-step interactions, long and noisy observations incur substantial context overhead and dilute the agent's attention to task-relevant information. Existing token pruning and summarization methods remain constrained by discrete text representations, limiting the achievable compression ratios. In this work, we challenge the prevailing assumption that...

    arxiv.org/abs/2611.06500 · PDF

  22. 22

    FPTR: Feasibility-Preserving Transactional Refinement for Deadline-Constrained Joint Beam and Resource Allocation

    Minghao Kong, Zirong Liu, Jianhao Wu, Muyao Chen, Jianxin Wu, Rongjie Wang

    cs.AI · cs.AR · cs.PF

    Short-interval wireless schedulers must return a complete, feasible beam-and-resource allocation before each execution deadline, and a late or invalid output gives no utility in the current control cycle. Searching longer can increase transmitted traffic, but it may also miss the deadline or leave only an infeasible intermediate solution. We propose Feasibility-Preserving Transactional Refinement (FPTR), which always keeps a validated...

    arxiv.org/abs/2611.06491 · PDF

  23. 23

    AmbiHTMLBench: A Benchmark for HTML-based Visually Rich Document parsing

    Boyu Niu, Yingjie Xiong, Shurui Li, Linhan Pang, Yuntong Zhang, Han Wang, Hao Wu, Kai Zuo, Xu Tang, Xuanhe Zhou, Fan Wu

    cs.AI

    Traditional document parsing systems use Markdown as their output representation, which is effective for linear text but does not fully preserve fine-grained spatial structure, table geometry, visual hierarchy, or rendering semantics. We show that HTML is advantageous for heterogeneous and layout-sensitive documents whose representations must remain editable, inspectable, and reusable, and introduce AmbiHTMLBench to systematically study this...

    arxiv.org/abs/2611.06449 · PDF

  24. 24

    SUMONE: A Simple and Efficient Summarizer for Plug-and-Play Agent Memory

    Jizhe Zhang, Ang Li, Chenheng Zhang, Yisen Wang

    cs.AI

    Long-horizon language-model agents accumulate increasingly large interaction histories, making context compaction essential for efficient execution. Existing learned compaction methods often rely on downstream task outcomes or solver behavior, coupling training to specific tasks or agents. We introduce SumOne, a standalone summarizer trained to preserve information in agent histories through masked reconstruction. Given a summary of the...

    arxiv.org/abs/2611.06445 · PDF

  25. 25

    EviSeek: An Evidence-Centric Agentic Framework for Embodied Active Information Seeking

    Jintao Li, Zhiyong Cao, Baihua Xiao, Yu Bai, Börje F. Karlsson

    cs.AI

    Active embodied question answering requires agents to acquire evidence through viewpoint changes and physical interaction. The central challenge is to avoid undirected interaction: taking plausible actions without specifying what evidence they should reveal for answering the question. Action plans and accumulated histories alone can leave the informational purpose of an interaction implicit. We introduce EviSeek, a training-free framework...

    arxiv.org/abs/2611.06434 · PDF

  26. 26

    D3D: Decoupled Diffusion Drafting for Asynchronous Speculative Decoding

    Baoyou Chen, Ruiqiao Mei, Guancong Lin, Tao Song, Weihao Yuan, Qun Liu, Siyu Zhu

    cs.AI · cs.LG

    Target-conditioned diffusion drafters predict candidate tokens in parallel, yet each new block must wait for updated target-model hidden states. This forces drafting and verification to run sequentially, limiting speedup even though each block is generated in one pass. We introduce D3D, a standalone diffusion drafter that removes this inter-round dependency. D3D combines blockwise denoising with auxiliary autoregressive supervision to achieve...

    arxiv.org/abs/2611.06425 · PDF

  27. 27

    GIST: Gibbs-Targeted Hierarchical Data Selection for Offline RLVR

    Xuliang Huang, Tianxiang Zhao, Chengwei Qin, Yao Shu

    cs.AI · cs.LG

    Offline reinforcement learning with verifiable rewards (RLVR) trains on a fixed pool of sampled responses and typically supervises every token of every response, although responses and tokens are not equally informative. We study which ones to supervise through the Gibbs-optimal policy of KL-regularized RLVR, which offline training targets, and measure the mismatch between a policy and this target with a bidirectional KL divergence. At the...

    arxiv.org/abs/2611.06395 · PDF

  28. 28

    MA-Mesh: Resilient Multi-Agent Collaboration via Role-based Self-Healing

    JunYong Lee, Woo Kyung Kim, Honguk Woo

    cs.AI

    Effective collaboration in multi-agent systems requires mechanisms for coordination and adaptability in dynamic and uncertain environments. However, existing multi-agent systems rarely model resilience explicitly, instead relying on ad-hoc output correction that limits systematic, runtime management of robustness. In this work, we propose Multi-Agent Mesh (MA-Mesh), a framework that constructs a role-based self-healing layer on top of a...

    arxiv.org/abs/2611.06389 · PDF

  29. 29

    LegalEvolver: Agentic Legal Judgment Prediction via Rubric-Guided Self-Evolution

    Xinyu Yang, Jiejun Tan, Chenlong Deng, Chenghao Zhang, Tongyu Wen, Zhicheng Dou

    cs.AI · cs.CL

    As LLM-based agents develop, they have been increasingly applied in the legal domain. Legal Judgment Prediction~(LJP) requires models to conduct multi-step reasoning and external legal evidence retrieval. Existing legal search agents' performance is often constrained by the quality of retrieval queries. They often formulate queries directly from the initial user request, overlooking latent legal issues and legally salient details. To address...

    arxiv.org/abs/2611.06381 · PDF

  30. 30

    Does Context Compression Preserve Refusal Alignment?

    Anthony Nguyen, Wenjun Lin

    cs.AI · cs.CL

    Context compression reduces inference cost by encoding inputs into compact representations while preserving semantic content. An open question is whether semantic preservation alone is sufficient to maintain downstream behaviours such as refusal alignment. We investigate this question and find that encoder-based compression systematically weakens refusal behaviour in instruction-tuned language models, despite high reconstruction fidelity....

    arxiv.org/abs/2611.06380 · PDF

  31. 31

    Corralling AI Risks with Grassroots Contracts

    Ehud Shapiro

    cs.AI · cs.CL · cs.DC · cs.MA

    AI's inflection point is a fountain of alarming risks. Digital platforms in particular are vulnerable to three major AI risks, as those in control of global platforms cede control to AI, of the algorithms, of membership and of content: (1) AI control of global platforms, social and economic, used by the majority of humanity, resulting in the majority of humanity becoming vulnerable to AI control. (2) Unaccountable AI agents operating...

    arxiv.org/abs/2611.06374 · PDF

  32. 32

    EnzymeRSI: A dual-loop recursive self-improvement system for enzyme design

    Genoria AI Team, Yuejie Hou, Lixiang Yang, Yisi Liu, Weiting Tang, Yueyuxiao Yang, Meng Yang

    cs.AI

    Improvement in enzyme engineering is usually pursued by optimising which candidates to assay within a fixed model and workflow; when the bottleneck arises from a mismatched method or a failing workflow, that route is difficult to escape. We present EnzymeRSI, a dual-loop system for recursive self-improvement (RSI). The outer loop evolves harness assets without changing model weights, the inner loop develops prediction and selection algorithms...

    arxiv.org/abs/2611.06321 · PDF

  33. 33

    STEK: Structural-Terrain-Enhanced Koopman-Inspired Cross-Channel Regularization for Retrospective Battery Multi-Target Estimation

    Jing Wang, Shiyuan Piao, Fugee Tsung

    cs.AI

    Retrospective battery characterization across heterogeneous cells is challenging because cycling records vary substantially with chemistry, protocol, cell format, capacity, and operating conditions. An endpoint charge--discharge curve captures the local condition of an evaluated cycle, while a cell-level lifecycle terrain summarizes structural evolution across the archived trajectory. We propose STEK, a structural-terrain-enhanced framework...

    arxiv.org/abs/2611.06299 · PDF

  34. 34

    Multi-Resident Data Association in Smart Homes via MDP Modeling and Behavior Cloning

    Xi Chen, Julien Cumin, Fano Ramparany, Dominique Vaufreydaz

    cs.AI · cs.NE

    Ambient sensors offer a privacy-preserving solution for activity recognition in smart homes, yet they face a critical challenge in multi-resident environments known as Data Association, which aims to identify the resident triggering a specific anonymous sensor event. Existing approaches fundamentally fail to maintain global consistency by often yielding conflicting associations for silent residents. Additionally, these methods frequently...

    arxiv.org/abs/2611.06296 · PDF

  35. 35

    DyPSI: Dynamic Physics Sensing via Joint Field and Sensor-Trajectory Generation

    Yizhou Zhang, Panqi Chen, Lei Cheng, Ting Zhang, Jianlong Li, Jianfeng Shi, Shikai Fang

    cs.AI

    Physics sensing in evolving environments requires reconstructing dense spatiotemporal fields from sparse observations while adapting where future observations should be collected. Existing reconstruction and sensor-placement methods have made substantial progress, but they typically rely on static sensing assumptions: they either use a single layout across all frames or optimize placements independently at each frame, which limits their...

    arxiv.org/abs/2611.06280 · PDF

  36. 36

    A Systematic Benchmark for Evaluating Multimodal Personalized Reasoning of Vision-Language Models

    Dohyun Kim, Junyong Lhim, Kwanyong Park

    cs.AI · cs.CV

    People often make decisions by holistically reasoning over heterogeneous forms of personal information. In this paper, we study the relatively underexplored capability of multimodal personalized reasoning in multimodal large language models (MLLMs). To this end, we introduce \texttt{MPRBench}, a novel and comprehensive benchmark specifically designed to evaluate personalized reasoning across a wide range of tasks and real-world challenges....

    arxiv.org/abs/2611.06259 · PDF

  37. 37

    HarnessFlow: MCTS-Guided Harness Optimization for LLM Agents

    Zhi Hong, Qianlong Wang, Hongchen Wei, Aidan Wu, Daohong Ying, Mingze Kong, Bei Liu, Chong Luo

    cs.AI

    The performance of large language model (LLM) agents depends on both the underlying model and the harness that governs interaction with tools and the environment. Yet harness design remains largely manual, and automating it requires navigating an open-ended search space through costly end-to-end evaluations. We introduce HarnessFlow, a framework for optimizing complete executable harnesses under a limited candidate-evaluation budget, with the...

    arxiv.org/abs/2611.06254 · PDF

  38. 38

    A Definition of Catastrophic Performance

    Henrik Marklund, Alex Infanger, Benjamin Van Roy

    cs.AI · cs.LG

    A central challenge in AI is specifying objectives that align with human preferences. While existing theoretical results elucidate how misspecification impacts performance, they typically do not distinguish between merely suboptimal and catastrophic outcomes. To enable rigorous analysis of catastrophic risk from AI misalignment, we introduce a formal definition of catastrophic performance. Our definition is based on two reference values. The...

    arxiv.org/abs/2611.06219 · PDF

  39. 39

    CARM: Conformal Adaptive Risk Masking for Reliable Scientific Super-Resolution

    Guangtao Zhang, Haonan Wu, Junxuan Gao, Zhengyang Liu, Zhaoxi Zeng, Zhihao Li, Jianguo Huang, Zhiqiang Liu

    cs.AI · cs.LG · math.PR

    Deep learning-based super-resolution (SR) can reconstruct fine-scale Earth-observation fields from coarse observations, but reconstruction accuracy alone does not provide a calibrated rule for identifying reliable spatial regions. Conformal Mask Calibration (CMC), built on conformal risk control, converts a spatial uncertainty score into a trust mask and controls the expected worst-case local distortion under exchangeability. However, its...

    arxiv.org/abs/2611.06151 · PDF

  40. 40

    SafeStart: Mitigating the Overlooked Safety Cost of Active Expert Reduction in MoE LLM Serving

    Erfan Baghaei Potraghloo, Saeid Shokoufa, Seyedarmin Azizi, Arya Fayyazi, Souvik Kundu, Mehdi Kamal, Massoud Pedram

    cs.AI · cs.LG

    Mixture-of-Experts (MoE) language models send each token through a few of many expert subnetworks. Serving methods make these models faster by running even fewer experts per token, and they are validated on accuracy benchmarks. We show that these methods can weaken the model's refusal of harmful requests, and we introduce SafeStart, which largely restores refusal while keeping most of the speedup. In vLLM on one RTX A6000, keeping about two...

    arxiv.org/abs/2611.06147 · PDF

  41. 41

    ModalSearcher: Weaving Visual Evidence into Autonomous Cross-Modal Deep Search

    Zhihan Yin, Ruyi Li, Chengyu Fang, Yifeng Yao, Nonghai Zhang, Pengfei Xiong, Terry Zhang, Huishuai Zhang, Dongyan Zhao

    cs.AI

    Static and incomplete parametric knowledge limits the performance of multimodal large language models on novel-knowledge and knowledge-intensive tasks, especially those requiring multi-hop evidence acquisition. Existing multimodal agentic search methods often fix visual evidence at the input or a predefined trajectory position, making it difficult for agents to select search modalities as their information needs evolve and to handle complex...

    arxiv.org/abs/2611.06131 · PDF

  42. 42

    From Handcrafted to Generative: RLVR-Optimized Icon Generation for Alexa+

    Rohan Sequeira, Matthew Rosenberg, Amin Mahmoudabadbozchelou, Oleg Poliannikov, Jonathan Rubin

    cs.AI · cs.LG

    Alexa+ devices with screens display animated icons during conversations to visually reinforce what is being said. The original set of these icons was hand-drawn by designers one at a time, a process that produces polished results but is time-consuming and cannot scale to the range of topics encountered in open-ended conversation. We describe an automated pipeline, deployed in production, that generates these vectorised animated icons from a...

    arxiv.org/abs/2611.06123 · PDF

  43. 43

    Beyond Risk Cues: A Mechanism-Level Benchmark and Verifiable Debate for Scam Video Detection on the Web

    Yifan Wang, Junyu Lu, Shun Zhang, Chaozhuo Li, Fanliang Bu, Rong Yin, Lingbin Bu, Changrong Min, Peng Chen

    cs.AI

    Online scam videos pose significant threats to individuals and society. Existing studies primarily focus on binary classification, offering limited insights into how scams work and the reasoning behind detection decisions. Moreover, they often overlook the challenge of distinguishing scam videos from superficially similar non-scam content, such as advertisements, and the robustness of detection models against misleading arguments. These...

    arxiv.org/abs/2611.06110 · PDF

  44. 44

    Reasoning-Trace Overflow in Code Generation: An Empirical Study of Explicit Reasoning Models

    Lukas Twist, Helen Yannakoudakis, Jie M. Zhang

    cs.AI · cs.SE

    Explicit reasoning can help language models generate code, but it also introduces a practical risk: the model may spend so much of its output budget reasoning that it never produces usable code. We study this behaviour in relatively small open-weight reasoning models, framing it as reasoning-trace overflow. Across seven models and six code benchmark settings, we measure how much of a fixed 32k-token output budget is spent on reasoning,...

    arxiv.org/abs/2611.06095 · PDF

  45. 45

    UniCompress: From Multi-LoRA Experts to One Compact Model for Edge Intelligence

    Mingkang Liu, Peipei Jiao, Weiming Meng, Tianyuan Liu, Lichen Pang, Junyan Li

    cs.AI

    Embodied and edge systems increasingly rely on vision-language models (VLMs) to support heterogeneous skills, yet deploying a separate task adapter for each skill while retaining a high-precision backbone creates substantial memory and management costs. We present UniCompress, a deployment-oriented framework that converts multi-task expertise into one compact, adapter-free model. Starting from a consolidated BF16 Qwen3-VL-4B teacher,...

    arxiv.org/abs/2611.06075 · PDF

  46. 46

    Which One Is Sounding? Binaural Sounding Instance Disambiguation for Audio-Visual Segmentation

    Haoze Ma, Yidi Li, Yu Ma, Pengyuan Zhao, Yi Shi, Bin Ren, Yanwen Guo, Dimitrios Kanoulas

    cs.AI · cs.SD

    Existing audio-visual segmentation methods mainly predict sound-related regions or category masks and often fail to identify the sounding instance when multiple same-class candidates coexist. Recent audio-visual instance segmentation methods address instance-level prediction but rely largely on motion and cross-frame association, which may become unreliable under sparse observations or inconspicuous motion. To this end, we introduce binaural...

    arxiv.org/abs/2611.06074 · PDF

This edition is part of The Daily Abstract — cs.AI archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.