cs.AI · 2026-11-09 · No. 168

Artificial Intelligence, 2026-11-09.

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

01 — The papers

62 entries
  1. 01

    CPU-ARC: Agentic Reasoning Caching for on-CPU Reasoning Reuse

    Peter Belcak, Greg Heinrich, Saurav Muralidharan, Shizhe Diao, Pavlo Molchanov

    cs.AI

    Large language models reinvent the wheel all the time. They conflate task-solving with the execution of the solutions, which forces them to solve every problem from scratch. We show that once a task has been reasoned through, the application of its solution to a new problem instance can be done very efficiently and with little involvement from language models. We introduce agentic reasoning caching, a method for constructing robust...

    arxiv.org/abs/2611.09891 · PDF

  2. 02

    On the Reliability of Majority-Vote Pseudo-Labels in Test-Time Reinforcement Learning

    Haoyu Wang, Ru Wang, Yuanyuan Wang, Jiaxian Guo, Mingming Gong

    cs.AI

    Test-time reinforcement learning (TTRL) adapts pretrained language models to unlabeled target-domain data using self-generated pseudo-labels, whose reliability remains poorly understood. We analyze majority-vote supervision for TTRL with online Direct Preference Optimization (DPO) and decompose its error into population-majority bias and prompt-dependent finite-rollout variation. Because population-majority bias is generally unidentifiable...

    arxiv.org/abs/2611.09872 · PDF

  3. 03

    Flow-Guided Target-Space Alignment via Path Consistency

    Ruizhi Yuan, Zeqiu Yu, Wei Gao, Wei Chen, Xiwei Tang, Lu Tang

    cs.AI · cs.CV · cs.LG

    Many multimodal alignment problems require mapping a source modality into a fixed space defined by a target modality, where the target space is not merely an intermediate embedding but is used directly for downstream analysis. In this setting, successful alignment should preserve both the global structure of the target distribution and the instance-level correspondence observed in paired source--target training data. Existing objectives...

    arxiv.org/abs/2611.09863 · PDF

  4. 04

    Ensemble Methodology for Reference-Free Confidence Estimation in LLM-based Code Summarization

    Marco Colombo, Balaji A Jayaram, Sowmyakumari Selvaraj, Sumit Tokkar, Animesh Kumar, Aravinda Prasad Sistla,...

    cs.AI · cs.PL

    This paper describes an approach and use case to estimate the correctness of LLM-generated COBOL summaries without access to human-authored references. The core intuition is that inter-model agreement on semantics is informative about truth, and that an LLM-as-judge can provide higher-fidelity pairwise assessments than traditional string- or embedding-based metrics. We operationalise this intuition with (i) controlled summary generation from...

    arxiv.org/abs/2611.09846 · PDF

  5. 05

    Can Agentic Systems Close the Gap in Radiology with Production AI Tools?

    Aubin Charley, Julien Khlaut, Charles Corbière, Pierre Manceron, Corentin Dancette, Léo Machado

    cs.AI · cs.LG

    Radiology AI has developed along two paths: specialized models that perform well on narrow tasks but cover few clinical findings, and generalist models that address a broader range of tasks but remain prone to hallucination and inaccurate measurements. Combining the two through agentic systems offers a route towards comprehensive radiology reporting, but published systems are tied to one modality or body part. We describe a radiology agentic...

    arxiv.org/abs/2611.09836 · PDF

  6. 06

    BitDecoding-FP4: Native FP4 KV Cache Decoding via Algorithm-System Co-Design

    Dayou Du, Woon Yee Ng, Guangming Sheng, Nick Lau, Jianyi Cheng, Luo Mai

    cs.AI · cs.AR · cs.CL · cs.PF

    The key-value (KV) cache dominates memory capacity and bandwidth in long-context LLM serving. While 4-bit microscaling formats (NVFP4, MXFP4) shrink the cache and run attention on FP4 Tensor Cores, a fundamental gap remains between algorithmic quantization and system-aware acceleration: existing FP4 KV systems use vanilla quantization that causes severe accuracy drops, whereas accuracy-oriented 4-bit algorithms rely on vector codebooks whose...

    arxiv.org/abs/2611.09825 · PDF

  7. 07

    Generative Latent Priors for Interpreting and Steering Protein Language Models

    Daniel Hinjos, Lucía Urcelay

    cs.AI · cs.LG

    Dominant methods for interpreting large models impose strong structural assumptions. Generative Latent Priors (GLP) are a complementary approach: a diffusion model that learns the distribution of a frozen model's activations directly, imposing no assumptions. We port GLP to ESM-2 activations (UniRef50) and evaluate it at two host scales (ESM-2-8M and -650M), both as a lens for protein representations and as a prior for editing them....

    arxiv.org/abs/2611.09818 · PDF

  8. 08

    Maya a1: Do the Options Read the Model, or Does Their Position? Order Sensitivity in a Non-Autoregressive Decision Model

    Mohammed Nayeem Khan, Logesh L

    cs.AI · cs.CL

    Laya answers typed multiple-choice questions in one forward pass by scoring a hidden state at one [MASK] marker per option. We re-engineer the checkpoint (421M parameters; a 2-layer option head with no positional tensor; a documented but unused parallel layout) and test whether the answer depends on where an option is listed. Presenting 951 Qwen-verified questions under every cyclic rotation of their options (or six random permutations for 12...

    arxiv.org/abs/2611.09803 · PDF

  9. 09

    Test-Time Compute Scales Better with Parallel Agent Teams

    Junlin Wang, Haozhe Chen, Ang Li, Rahul Chalamala, Zhuang Liu

    cs.AI

    Test-time scaling improves the performance of a fixed model by spending more computation at inference. There are two common ways to spend it on agents when a verifier can score every attempt. Best-of-N sampling runs many independent agents and keeps the best result but saturates quickly. One large team of collaborating agents can score higher but stops improving as it grows. Can we get the best of both? Our method, parallel agent teams, does...

    arxiv.org/abs/2611.09798 · PDF

  10. 10

    Knowing When to Intervene: Causal Evaluation of AI Agents

    Ayush Sawarni, Wenqian Xing, Tijana Zrnic, Ramesh Johari, Vasilis Syrgkanis

    cs.AI · cs.LG · econ.EM · stat.ME

    Current evaluations of agentic AI systems primarily rely on benchmarks. However, the static nature of benchmarks makes them uninformative about the value of auditing interventions at deployment time. For reliable deployment, we often need to determine within an agent's trajectory whether or not to intervene, e.g., by verifying the agent's current output or steering it toward different behavior, using either human review or more costly models...

    arxiv.org/abs/2611.09792 · PDF

  11. 11

    Assessing the Impact of Retrieval-Augmented Generation on Algorithmic Problem-Solving

    Gabriel Souza Baggio, Gabriel Machado Lunardi, Guilherme Medeiros Machado, Subhankar Maity, Faiza Belbachir, José...

    cs.AI · cs.HC

    Competitive programming problems pose distinctive challenges for Large Language Models, as they demand not only proficient code generation but also advanced algorithmic reasoning. This study presents a Retrieval-Augmented Generation approach specifically designed to enhance the capacity of Large Language Models to address such problems. Our approach integrates an embedding-based retrieval mechanism and a similarity search algorithm to...

    arxiv.org/abs/2611.09791 · PDF

  12. 12

    Continuous wavelet-based feature extraction for laser-induced breakdown spectroscopy classification of human plasma samples

    Md Mahmudur Rahman, Atqiya Munawara Mahi, Nazim Ahmed Belabbaci, Mohammad Arif Ul Alam

    cs.AI

    Extracting discriminative features from laser-induced breakdown spectroscopy (LIBS) spectra is important for the classification of complex biological samples. This study evaluates continuous wavelet transform (CWT)-based spectral peak extraction for distinguishing human plasma samples labeled as SARS-CoV-2-positive or negative. Plasma specimens from 181 individuals, comprising 53 positive samples and 128 negative controls, were deposited as 5...

    arxiv.org/abs/2611.09746 · PDF

  13. 13

    What to Delegate, What to Preserve: Human Preferences for Agentic AI Across Everyday Life and Work

    Sarah Wheeler, Veronica Rammouz, Yaoyao Qian, Zexuan Liu, Mengyi Wei, Anthony Rios, Tianzhi He, Kyrie Zhou

    cs.AI · cs.HC

    AI agents can execute multi-step tasks on users' behalf, yet the boundaries of acceptable delegation remain unclear. We report a scenario-based survey of 289 U.S. adults with prior AI experience covering 45 use cases across nine life domains. Each respondent evaluated one scenario per domain, producing repeated measures of automation preference, trust, collaboration, AI leadership, privacy concern, and willingness to pay. Automation...

    arxiv.org/abs/2611.09730 · PDF

  14. 14

    FOSCAR: Accelerating Decode Attention over 2- and 4-Bit KV Caches through Affine Folding

    Donglin Zhuang, Pragaash Ponnusamy, Xiaoxia Wu, Shuaiwen Leon Song, Zhongzhu Zhou

    cs.AI

    Large-language-model decode attention streams the key-value (KV) cache for every generated token. Packing that cache into 2-bit or 4-bit quantized integers reduces memory traffic but does not automatically translate into latency saving. Existing 2-bit fused kernels reach under 31%/15% of peak HBM throughput on H100/B200. Converting each value and applying its scale and offset before feeding Tensor Cores is a major bottleneck because...

    arxiv.org/abs/2611.09722 · PDF

  15. 15

    Intent-Aware Caching for Efficient LLM Serving

    Kazi Hasan Ibn Arif, JinYi Yoon, Dimitrios S. Nikolopoulos, Hans Vandierendonck, Deepu John, Bo Ji

    cs.AI · cs.LG

    Modern LLM serving systems use prefix caching to accelerate multi-turn conversational workloads. They cache the key-value states of previous conversation prefixes and reuse them when subsequent requests share the same prefix, to avoid redundant prefill computation and reduce serving latency. However, achieving a high cache hit rate remains challenging because cache reuse depends on whether a conversation continues and how soon the next turn...

    arxiv.org/abs/2611.09715 · PDF

  16. 16

    Attend at the Right Resolution: Head-Adaptive Sparse Attention via Query-Key Interaction Structure

    Wajih Hassan Raza, Ahsan Bilal, Yidan Shen, Renjie Hu, Xin Fu, Yu Wen

    cs.AI

    Attention is a major scalability bottleneck in large language models (LLMs), with its computational cost and key-value (KV) memory traffic becoming increasingly prohibitive as context length grows. Sparse attention addresses this bottleneck by restricting each query to a subset of the context, but its effectiveness hinges on identifying relevant regions both cheaply and reliably. Existing training-free methods primarily rely on runtime...

    arxiv.org/abs/2611.09700 · PDF

  17. 17

    Not Your Diary, Just Your Topic: Grading the Wrong-User Control

    Karina Romanova, Igor Kiselev, Vladimir Platonov, Viliana Devbunova, Gregory Leleytner, Adeliya Leleytner, Lina...

    cs.AI

    A common test of personalization replaces a user's history with someone else's and measures the drop in accuracy. But this swap also changes the topics in the history, so the drop may reflect topic mismatch. We examine this problem with a frozen language model on LaMP-2 news categorization, using histories from similar users and synthetic histories assembled from other users' articles. The user's own history improves accuracy by 2.3...

    arxiv.org/abs/2611.09609 · PDF

  18. 18

    Alignment by Confident Debate: Preference Learning from Multi-Agent Debate

    Nikhil Verma, Dingxian Wang, Mohammad Reza Rezaei, Andrew Rabinovich, Zhao Chen

    cs.AI

    Post-training language models for reasoning typically rely on gold answers, task-specific verifiers, or learned reward models, limiting scalability when such signals are scarce. Multi-agent debate offers a self-supervised alternative, but converting debate into reliable training supervision remains challenging: majority voting weights agreeing responses equally, lacks a principled tie-breaking rule, and is ill-defined for open-ended tasks...

    arxiv.org/abs/2611.09568 · PDF

  19. 19

    CoMoFlow: Controllable Protein Conformational Ensemble Generation via Explicit Collective-Mode Modeling

    Yusong Wang, Zheyuan Gu, Ailiang Lin, Minghao Shao, Yicheng Xu, Jiang Cai, Prayag Tiwari, Mingkun Xu

    cs.AI · cs.LG

    Protein conformational ensemble generation aims to efficiently characterize the structural variations accessible to a protein beyond a single static structure. Recent flow-based methods efficiently generate conformational ensembles, but they model coordinated structural variation implicitly, which can limit ensemble fidelity and hinder direct control. To address this limitation, we propose CoMoFlow, a controllable framework for generating...

    arxiv.org/abs/2611.09548 · PDF

  20. 20

    CID: Measuring Cognitive Work Through Semantic State Change

    Ron Fulbright

    cs.AI

    Traditional information theory provides powerful tools for analyzing statistical properties of information but offers limited insight into meaning and semantic transformation. It has never been possible to quantify the meaning of an arbitrary piece of information. Large language models represent information as vectors embedded within high-dimensional semantic spaces, enabling meaning to be analyzed geometrically. This paper defines semantic...

    arxiv.org/abs/2611.09542 · PDF

  21. 21

    PreForceVision: Physics-Guided Optical Estimation of Bolt Preload from Bolt-Head Deformation

    Nehal Afifi, Klaus Rappenecker, Florian Peinhardt, Gerhard Kachel, Alfred Isele, Sven Matthiesen

    cs.AI

    Bolt preload is a decisive functional quantity in bolted joints because it determines whether external loads are transmitted primarily through frictional clamping or directly through the bolt. Existing preload measurement techniques, including strain-gauge bolts, force washers, ultrasonic methods, and bolt-length measurements, are effective in certain applications but require additional hardware, coupling, calibration, or joint modifications....

    arxiv.org/abs/2611.09529 · PDF

  22. 22

    StockASkill: A Local-First AgentSkill for Intelligent Stock Selection

    Xiangjing An, Zhibin Jiao, Aiting Chen

    cs.AI

    Quantitative equity selection and portfolio construction in emerging and cross-border markets frequently suffer from computational bottlenecks, remote API fragility, survivorship bias, and rigid execution constraints such as discrete board lots and investor-mandated drawdown limits. In this paper, we present StockASkill, a local-first, full-data-driven AgentSkill engineered as a specialized Quant Co-Processor for autonomous AI agents and...

    arxiv.org/abs/2611.09528 · PDF

  23. 23

    Atik: A Shared-Mesh Transformer Accelerator for RISC-V

    Ahmed Zeer, Erkan Uslu

    cs.AI · cs.AR

    Transformer inference repeatedly invokes dense matrix multiplication and attention, including its data-dependent softmax normalization. We present Atik, an open-source RoCC accelerator for RocketChip that executes matrix multiplication and dense attention through one shared fixed-point MAC mesh. Atik accepts and returns BF16 tensors, uses a streaming online-softmax schedule that never materializes the full score matrix, and is invoked through...

    arxiv.org/abs/2611.09522 · PDF

  24. 24

    A Physics-Informed Learning Framework for Linking Microscopic Surface Degradation to Bolted Joint Functional Behavior

    Nehal Afifi, Gerhard Kachel, Sven Matthiesen

    cs.AI

    Returned products in circular engineering often have heterogeneous degradation histories, making remaining function difficult to assess. For bolted joints, corrosion alters surface morphology, contact, friction, preload retention, and mechanical response. This paper presents a physics-informed multimodal framework for linking microscopic and camera-based degradation observations to functional indicators of corroded bolted joints. The workflow...

    arxiv.org/abs/2611.09514 · PDF

  25. 25

    LEMMO: A Large Electromagnetic Model leveraging Pretrained LLMs for Million-Sample Signal Understanding

    Mingkai Li, Jiajun Wu, Yihan Mao, Zilin Sang, Taosong Yu, Yuan Ma, Minghe Liu, Qianqian Yang, Zhiguo Shi

    cs.AI · cs.SD

    Large language models have expanded beyond text to vision, audio, and other modalities, but understanding raw electromagnetic (EM) signals remains challenging. Wideband recordings can contain millions of samples, whereas existing waveform-based EM models are primarily evaluated on short in-phase and quadrature (IQ) segments, leaving long-recording understanding and detailed description insufficiently explored. We introduce LEMMO, a large EM...

    arxiv.org/abs/2611.09476 · PDF

  26. 26

    ENGRAM: From Agent Experience to Weights through Verified Consolidation

    Ruiyi Yang, Zechen Li, Lihuan Li, Du Yin, Hao Xue, Flora D. Salim

    cs.AI · cs.IR

    Language-model agents use experience through external retrieval and parameter learning. Consolidating stored experience into model weights requires deciding both which experience to train on and when its external records can leave retrieval. Admission based solely on student mastery can exclude useful teacher demonstrations, while removal immediately after training can discard guidance that the model still needs. We introduce ENGRAM, a...

    arxiv.org/abs/2611.09472 · PDF

  27. 27

    Definitions, Label Names, or Demonstrations? Auditing What Governs LLM Annotation

    Viraj Mishra, Shubham Gupta, Dhruv Gupta, Yuehong Cassandra Tai, Qunhua Li

    cs.AI

    Large language models (LLMs) are increasingly used to annotate social-science texts, but high accuracy does not establish that they apply researcher-supplied category definitions: predictions may instead reflect prior associations with category labels or labeled demonstrations. We introduce a counterfactual audit that reassigns definitions among output labels, distinguishing adherence to supplied definitions from preservation of a dataset's...

    arxiv.org/abs/2611.09457 · PDF

  28. 28

    Agentic AI for Healthcare Claim Denial Analysis: A Policy-Grounded Framework for Privacy Preservation

    Sofia Campos, Carlos Martins, Pedro Pires, João Gama, Bruno Veloso, Pedro Santos

    cs.AI · cs.LG

    In the healthcare sector, claim denials impose significant financial and administrative burdens, emphasizing the need for automated and trustworthy decision-support systems. This paper proposes a modular Agentic Artificial Intelligence (AI) framework for healthcare claim denial analysis, integrating specialized agents for denial risk prediction, risk factor identification, diagnosis-procedure-medication code relationship analysis, and...

    arxiv.org/abs/2611.09423 · PDF

  29. 29

    Typed-Centrality Grounding for Agentic GraphRAG: A Symbolic Approach for Agentic Reasoning

    Rosario Napoli, Antonio Celesti, Massimo Villari, Maria Fazio

    cs.AI

    Knowledge Graphs (KGs) and Large Language Models (LLMs) represent two approaches for knowledge representation and reasoning over heterogeneous data. Recent research has explored their integration through Graph Retrieval-Augmented Generation (GraphRAG) systems, which use KGs to provide structured context to LLMs. However, existing GraphRAG approaches treat KGs as homogeneous graphs and rely on textual similarity between the input prompt and...

    arxiv.org/abs/2611.09414 · PDF

  30. 30

    MIRA: Mitigating Pole Geometry Bias in Recurrent Language Models

    Róisín Luo, Ihsan Ullah, Christian Gagné, Dong Li, Karyn Morrissey

    cs.AI

    We show that the Transformer's recurrent successors --- such as selective state-space models (SSMs), Receptance Weighted Key Value (RWKV), and Retentive Network (RetNet) --- can be characterized as infinite-impulse-response (IIR) systems. The zero--pole geometries of these IIR systems characterize the temporal modes they can represent. However, to trade parallelism, the algebraic structures of the parameterizations of these recurrent systems...

    arxiv.org/abs/2611.09412 · PDF

  31. 31

    ROAM: Retrieval-Augmented Optimization for Feasible Multi-Day Itinerary Planning

    Ye Zihao, Lin Xiaoheng

    cs.AI

    Large language models (LLMs) can draft plausible travel itineraries, but the plans they produce are frequently infeasible: attractions are grouped into days without regard for geography, opening hours are ignored, and a single generation pass cannot reliably organize a multi-day trip. We present ROAM, a retrieval-augmented agentic framework for spontaneous multi-day itinerary planning that couples semantic preference retrieval with explicit...

    arxiv.org/abs/2611.09407 · PDF

  32. 32

    PhyCAD: Physics-Aware Text-to-CAD Generation via Hierarchical Reinforcement Learning

    Yuxin Cui, Weijie Zheng, Zhengyang Li, Han Wang, Wenjie Feng, Qiyu Kang

    cs.AI

    Language-driven computer-aided design (CAD) generation enables the creation of editable parametric models from text, yet geometric correctness alone does not ensure physical feasibility. We present PhyCAD, a framework for physics-aware Text-to-CAD generation that trains language models to produce executable CadQuery programs under geometric and physical constraints. To support this setting, we construct a Text-to-CAD dataset that links...

    arxiv.org/abs/2611.09401 · PDF

  33. 33

    Travel Cost Calibration for Multi-Agent Target Assignment and Pathfinding

    Yu Kumagai, Ryohei Banno, Keisuke Okumura

    cs.AI

    Target assignment and pathfinding (TAPF) is an appealing extension of classical multi-agent pathfinding (MAPF): each agent has a set of targets, and the planner must both assign a target to each agent and plan collision-free paths for them. A trivial approach is to compute an assignment that minimizes the total single-agent shortest-path distance over agent--target pairs, and then solve the resulting MAPF instance. However, this approach...

    arxiv.org/abs/2611.09396 · PDF

  34. 34

    Against All Likelihood

    David Lopez-Paz

    cs.AI

    Maximum likelihood is the dominant training principle for large language models, yet likelihood is neither necessary nor sufficient for truth. Unlikely events happen every day, and frequently repeated statements can be false. Drawing on a philosophical tradition that ties objectivity to invariance rather than probability (from Nozick's account of objective facts and Wigner's principles of invariance to the structural realism of Poincaré and...

    arxiv.org/abs/2611.09385 · PDF

  35. 35

    Exploring Weight-Space Programming: Capabilities, Generalization, and Limitations

    Junru Hu, Cheng Tang, Wei Li, Hongze Zhu, Chun-Mei Feng, Qi Zhang, Yirong Chen, Ming Hu

    cs.AI · cs.LG

    Weight-Space Programming (WSP) enables reasoning agents to directly inspect and modify neural network parameters, offering an alternative to gradient-based model adaptation. However, its capabilities, cross-task generalization, and practical limitations remain insufficiently understood. We systematically investigate WSP through 16 editing trajectories across seven benchmarks and two model scales (Qwen3.5-0.8B and Qwen3.5-9B). Our findings...

    arxiv.org/abs/2611.09382 · PDF

  36. 36

    KineCouple: Substrate-conditioned residual readout for enzyme kinetic prediction

    Chun Kang, Yunbo Zhang, Fanhu Zeng, Jian Liu, Siwei Deng, Lei Sha

    cs.AI · q-bio.QM

    Accurate prediction of enzyme kinetic parameters can support enzyme engineering and metabolic modeling by complementing experimental characterization. Existing prediction methods often pool residue-level protein features before combining them with substrate features. Such substrate-independent pooling cannot adapt residue weighting to the substrate during aggregation, potentially diluting information relevant to a specific enzyme-substrate...

    arxiv.org/abs/2611.09377 · PDF

  37. 37

    Enabling Privacy-Preserving Natural Language Access to Industrial Smart Meter Data

    Carlos Martín-Abellán, Mark Kröll

    cs.AI · cs.CL · cs.DB · cs.MA

    Energy management officers and grid operators increasingly rely on smart meter data for anomaly detection, performance monitoring, and strategic decision-making. However, extracting actionable insights from large-scale, heterogeneous telemetry databases requires advanced SQL and data engineering expertise, creating a critical bottleneck for non-technical personnel. Furthermore, the strict data privacy regulations and security constraints...

    arxiv.org/abs/2611.09372 · PDF

  38. 38

    SafeOrchestrate: Runtime Checked Intent-Driven Orchestration for Cyber-Physical Systems

    Muhammad Ashfaq, Ahmed R. Sadik, Ewan Warburton, Pyry Kotilainen, Niko Mäkitalo, Tommi Mikkonen

    cs.AI

    Modern cyber physical systems (CPS) integrate heterogeneous sensors, actuators, controllers, and services, composed at runtime rather than at design time. Orchestrating such systems through natural language (NL) is attractive, but unconstrained large language model (LLM) tool use is risky because actions must respect device capabilities, runtime state, and safety constraints. We present Safe Orchestrate, a reference architecture for runtime...

    arxiv.org/abs/2611.09364 · PDF

  39. 39

    Challenges, Opportunities, and Misconceptions in Neurosymbolic AI

    Giuseppe Marra, Pierre Genevès, Meghyn Bienvenu, Thomas Schiex, Sylvie Thiébaux, Silviu Maniu

    cs.AI

    We examine opportunities for neurosymbolic AI alongside increasingly powerful neural models, particularly with regards to efficiency, safety and interpretability. We identify two major challenges: ensuring semantic alignment between neural and symbolic components, and managing the computational cost of symbolic inference and learning. We discuss representative methodological approaches to these challenges and address several common...

    arxiv.org/abs/2611.09356 · PDF

  40. 40

    AUTEUR: Acoustic Plan Learning for Red-Teaming Text-to-Audio Models

    Jiashu Zhang, Jiaming He, Zheming Zhang, Kaizheng Wang, Bing Li, Zhiyuan Zha, Yi Yu

    cs.AI · cs.SD

    Automated red-teaming of text-to-audio models requires test prompts that pass textual screening and expose harmful acoustic content. The central challenge is to connect the sounds produced by the target model to both reusable test generation and prompt revision. Our key idea is to represent test descriptions as structured acoustic plans that organize foreground sounds, context, and vocal attributes. To this end, we propose acoustic plan...

    arxiv.org/abs/2611.09351 · PDF

  41. 41

    ABHBench: Evaluating Moral Decision-Making of Foundation Models from an Agentic Perspective

    Chaoran Li, Yukun Li, Zeyuan Zhao, Shao Zhang, Xihuai Wang, Ying Wen

    cs.AI · cs.CL · cs.CY

    Foundation models are increasingly deployed as agentic systems that perceive context and act autonomously, raising the need to evaluate their moral behavior as active decision-makers. Existing benchmarks, however, largely ask models to judge human actions from third-person textual vignettes, leaving untested how models behave when they themselves become the acting agent. We introduce ABHBench, a multimodal benchmark built from Detroit: Become...

    arxiv.org/abs/2611.09326 · PDF

  42. 42

    Towards an Autonomous AI Steward for Large-Scale Physics Experiments: Integrating Causal Diagnostics, RAG, and Conformal Prediction

    Oliver Noll, Nicholas Tan Jerome, Alexander Lehnen, Frank Simon, Frank Maas

    cs.AI

    Long-term particle physics experiments face critical operational challenges: monitoring non-stationary machine data streams and preserving tacit expert knowledge. We present the "Helmholtz Co-Pilot," a hybrid Multi-Agent System (MAS) acting as an autonomous experiment steward. The system can be triggered interactively by human operators or autonomously by an asynchronous causal diagnostics engine that runs independently of the MAS reasoning...

    arxiv.org/abs/2611.09298 · PDF

  43. 43

    AutoRefine: Auto-Calibrating Code Judges from Workflow Outcomes

    Shervin Shahidi, James Poon, Hang Xie, Kenneth O. Xu, Robin Håkanson, Vikram Nair, Jens Pillgram-Larsen, Nachiappan Nagappan

    cs.AI · cs.SE

    LLM judges increasingly serve as quality gates for code changes that have no test oracle, but a judge prompt must be calibrated for its task and recalibrated as the system it judges evolves, while hand labels for the calibration are resource intensive and go stale. Development workflows emit outcome labels for free (tests pass or fail, changes land or are abandoned), but these labels are noisy. We present AutoRefine, a loop that rewrites a...

    arxiv.org/abs/2611.09291 · PDF

  44. 44

    Ontology-Driven Control of Conversational Agents with a Use-Case for Job Interview Practice

    Barbara Gendron, Gaël Guibon, Mathieu D'aquin

    cs.AI

    Large Language Models (LLMs) have been widely used as conversational agents in both domain-specific and open-domain situations. In domain-specific context, it is often necessary to adapt those LLMs to obtain conversations with desired characteristics, whether those apply at the level of each response (e.g. for tone or style) or more globally (i.e. at the level of a conversational strategy). Such adaptations are, however, generally realized in...

    arxiv.org/abs/2611.09282 · PDF

  45. 45

    An Empirical Study of Issues in GPU Kernel Libraries for LLM Systems

    Yanjie Gao, Jiayi Bi, Haoxiang Lin, Ling Feng

    cs.AI

    Behind the expanding use of large language models (LLMs) across application domains lies a growing dependence on specialized GPU kernel libraries. These libraries underpin the correctness and efficiency of training and inference across diverse GPU platforms, implementing attention, GEMM, quantization, and low-precision operators while managing tensor shapes, memory layouts, code generation, and hardware-specific dispatch. Problems at this...

    arxiv.org/abs/2611.09280 · PDF

  46. 46

    PHISHREV+: An ASP-Based Post-Hoc Defeasible Machine Learning Framework for Phishing Detection

    Mainak Sen, Kumar Sanka Ray, Amlan Chakrabarti

    cs.AI · cs.LG

    Phishing website detection remains a critical cybersecurity challenge. Although machine learning classifiers achieve high accuracy on benchmark datasets, they suffer from a fundamental limitation: once trained, their decision boundaries are fixed and cannot incorporate new contextual evidence without retraining the model. This work presents \textbf{PHISHREV+}, a hybrid framework that instantiates the notion of \textit{post-hoc defeasible...

    arxiv.org/abs/2611.09273 · PDF

  47. 47

    Zero-Shot Weight-Space Programming Across a Difficulty-Graded Ladder of Two-Player Games

    Eduardo C. Garrido-Merchán

    cs.AI

    Weight-space programming sets every weight of a feed-forward network by hand, from an explicit specification of the desired computation, with no gradient descent and no training data anywhere in the construction. We ask how far a large language model can carry this practice, zero-shot, across games of deliberately increasing difficulty. We design a four-rung ladder (Nim, tic-tac-toe, two sizes of Connect Four, and a chess endgame) spanning...

    arxiv.org/abs/2611.09271 · PDF

  48. 48

    Nudging Decision Models: Do Digital Nudges Steer Fast, Generation-Free AI Agents?

    Sascha Kaltenpoth, Haya Halimeh, Kevin Bösch, Oliver Müller

    cs.AI

    LLM-based agents increasingly automate business and consumer tasks but are slow and expensive. Decision models, a new class that chooses among options without generating text, are far faster and less costly. Developers call them "System One models," referencing dual process theory. We ask whether such models can be steered by digital nudges online and whether, as the theory predicts, automatic nudges affect them more than reflective ones. We...

    arxiv.org/abs/2611.09264 · PDF

  49. 49

    Model-based learning for DoA estimation and array calibration via continuous sparse recovery

    Baptiste Chatelier, José Miguel Mateos-Ramos, Vincent Corlay, Matthieu Crussière, Christian Häger, Henk Wymeersch,...

    cs.AI · cs.LG

    Direction of arrival (DoA) estimation, a classical signal processing problem, has experienced renewed interest with the advent of integrated sensing and communication (ISAC). While decades of research have yielded highly accurate DoA estimators, hardware impairments have frequently been neglected, despite degrading both DoA estimation accuracy and downstream communication performance in ISAC systems. This work introduces a joint array...

    arxiv.org/abs/2611.09263 · PDF

  50. 50

    Forecasting cumulative intraoperative hypotension burden: patient-disjoint evaluation of quantile boosting and alert policies

    Yunfan Zhao

    cs.AI

    Background: Binary hypotension forecasts do not describe the depth and duration of future low pressure. Burden forecasts may support graded early warning, but better forecast scores need not yield more useful alerts. Objective: To evaluate conditional quantiles of the next 15-minute mean arterial pressure (MAP) deficit below 65 mmHg and distinguish forecasting performance from episode-warning performance. Methods: Public VitalDB numeric...

    arxiv.org/abs/2611.09259 · PDF

  51. 51

    Zero-Query Transferable Pretend Benign Attacks with Amortized Online Generation for Cooperative Perception

    Hongwei Lin, Xun Huang, Chenglu Wen, Cheng Wang

    cs.AI

    Cooperative perception enables an ego agent to acquire information from collaborators beyond its own sensors' reach. Yet, the most beneficial messages are often the hardest to verify from local observations. Pretend Benign (PB) exploits this asymmetry by targeting regions with weak local verifiability and applying multi-objective perturbations, making harmful feature changes appear to be benign assistance. However, PB is victim-dependent:...

    arxiv.org/abs/2611.09256 · PDF

  52. 52

    MM-DyGraph: A Dataset, Benchmark, and Model for Multimodal Dynamic Graphs

    Zongyuan Wu, Pengjie Wang, Lingshan Chen, Tianhang Wan, Yuan Meng, Xin Wang, Ling Feng, Wenwu Zhu

    cs.AI · cs.LG

    Multimodal dynamic graphs are pervasive in real-world applications, where nodes and edges are accompanied by multimodal signals, including text and images. Despite their practical importance, benchmarks and methods that explicitly study multimodal learning in dynamic graphs remain scarce, hindering the full understanding and effective utilization of multimodal information for downstream tasks. To address this gap, we introduce MM-DyGraph, to...

    arxiv.org/abs/2611.09240 · PDF

  53. 53

    RAG-Enhanced LLMs for Interactive Explainability in Clinical Decision Support Systems

    Lorenzo Mannocci, Francesca Naretto, Lucia Passaro, Anna Monreale

    cs.AI · cs.HC · cs.LG

    Artificial Intelligence (AI) is increasingly embedded in high-risks decision-making systems across domains such as healthcare, finance, and the judicial system. As reliance on predictive models grows, so does the demand for Explainable AI (XAI) to ensure transparency, trust, and interpretability. However, existing XAI approaches often fail to present explanations in a format accessible to domain experts, limiting their practical utility....

    arxiv.org/abs/2611.09232 · PDF

  54. 54

    Fused-Softmax for efficient quantized attention

    Tianxing Wu, Juuso Korhonen, Jesus Carabano Bravo

    cs.AI · cs.LG · cs.PF

    Low-precision matrix throughput on modern GPUs is growing far faster than vector-arithmetic throughput. Attention therefore cannot realize the available GEMM performance because softmax has become the pipeline bottleneck. We introduce Fused-Softmax Attention, a training-free algorithm that reformulates and approximates softmax to fold its dominant operations into the attention GEMMs, moving work from vector units to matrix cores. Implemented...

    arxiv.org/abs/2611.09225 · PDF

  55. 55

    Safe Reinforcement Learning via Probabilistic Option Shielding

    Gian Luca Ascari, Isidoro Tamassia, Ying Jiao, Luc De Raedt, Giuseppe Marra

    cs.AI · cs.LG

    Shielding is a safe reinforcement learning technique that provides safety guarantees while learning, by blocking unsafe actions. Traditional shielding relies on a complete model of the environment to provide long-term guarantees. Probabilistic logic shielding (PLS) lifts the requirement of having such model, but only considers safety in the short term, typically looking only one or few steps ahead. We propose Probabilistic Option Shielding...

    arxiv.org/abs/2611.09197 · PDF

  56. 56

    Large Language Model Consensus and Corpus-Adaptive Topic Modeling for Classifying and Mapping Patent Innovation

    Sercan Ozcan, Enes Ozgozler, C. Okan Sakar

    cs.AI · cs.IR

    Patent analytics increasingly relies on transformer representations and large language models (LLMs), yet linking conceptual classification, topic formation and interpretation requires explicit decisions about model roles and the evidence retained at each stage. In this design-oriented study, we propose and demonstrate a patent analytics pipeline that links these tasks and keeps each decision open to audit. Three LLMs separately classify each...

    arxiv.org/abs/2611.09194 · PDF

  57. 57

    Analyzing Code Provenance in LLMs for Programming Education

    Subhankar Maity, Faiza Belbachir, Omar Boucelma, Guilherme Medeiros Machado

    cs.AI · cs.SE

    Large language models (LLMs) play an increasing role in programming education, challenging traditional assessment methods as they can generate fully functional code instantly. This disrupts the reliability of student submissions as indicators of learning and effort. To address this issue, code provenance is essential for tracing the origins of code and distinguishing student-authored, AI-generated, and externally sourced components. We extend...

    arxiv.org/abs/2611.09177 · PDF

  58. 58

    TailDART: Compute-Efficient Foundation Models for Financial Time-Series Forecasting

    Chiang Qin Zhi, Ranjan Satapathy, Erik Cambria, Joyjit Chattoraj

    cs.AI · cs.CE

    Although time-series foundation models (TSFMs) achieve strong zero-shot transferability across high-information-density physical systems, applying overparameterized TSFMs to quantitative finance is fundamentally constrained by low signal-to-noise ratios, non-stationarity, and prohibitive autoregressive decoding latency. To reconcile representation learning with institutional operational constraints, this study presents TailDART, a dense 2.68...

    arxiv.org/abs/2611.09176 · PDF

  59. 59

    Evaluation of a Multi-Link Machine Learning Approach for Rainfall Detection Using Satellite Microwave Links

    Mostafa Ftouni, Margherita Spalla, Matteo Colli, Andrea Randazzo, Christian Gianoglio

    cs.AI · cs.LG · eess.SP

    The impacts of climate change have increased the need for reliable real-time rainfall monitoring systems. This study investigates rainfall detection using a network of opportunistic sensors based on satellite microwave links (SMLs) combined with machine learning (ML) classifiers. Two models, a random forest (RF) and a convolutional neural network (CNN), were trained on aggregated SML data, including information from neighboring links. The...

    arxiv.org/abs/2611.09156 · PDF

  60. 60

    Learning When to Think: Dual-Reference Offline Optimization for Adaptive VLM Reasoning

    Hongbin Lin, Sizhe Zou, Juangui Xu, Xinyue Xu, Zhuoli Ouyang, Stanislav Kolchin, Chengwei Qin, Zhongxiang Dai, Yao Shu

    cs.AI

    Reasoning traces improve vision-language models on complex multimodal tasks, but forcing every input to follow a long reasoning process leads to overthinking and unnecessary inference cost. We study adaptive fast/slow thinking for VLMs, where one model answers directly for simple inputs and reasons step by step when needed. Fast answers and slow traces often come from distinct behavior policies, making a single-reference offline objective...

    arxiv.org/abs/2611.09154 · PDF

  61. 61

    BMMS: Policy-Conditioned Measurement of Behavioral Compliance in Language Models

    Anthony Rhodes, Gadi Singer

    cs.AI · cs.CL

    Behavioral evaluation of language models fixes its dimensions in advance: hallucination, sycophancy, toxicity, over-refusal. An operator in a regulated industry needs a different question answered: does a system do what a particular rulebook requires? We introduce the Behavioral Misalignment Metric Suite (BMMS), which measures policy-conditioned misalignment by taking a policy corpus as an argument and returning a compliance profile over it,...

    arxiv.org/abs/2611.09147 · PDF

  62. 62

    CARGO: Causal Attribution and Responsibility-Guided Preventive Safe Multi-Agent Reinforcement Learning for Cooperative Driving

    Jingjia Teng, Yang Li, Wenjie Huang

    cs.AI

    This study presents Causal Attribution and Responsibility-Guided Optimization (CARGO), a multi-agent reinforcement learning framework for cooperative on-ramp merging in mixed traffic. A shared latent-context posterior and an action-conditioned response model estimate directed behavioral influence, with posterior uncertainty propagated into interaction-edge selection. Complementary attribution mechanisms couple signed influence with marginal...

    arxiv.org/abs/2611.09139 · 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.