cs.AI · 2026-07-25 · No. 64

Artificial Intelligence, 2026-07-25.

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

01 — The papers

64 entries
  1. 01

    Unsupervised Consensus-Based Anomaly Detection for Spatiotemporal Malaria Incidence in Ghana

    T. Ansah-Narh, Y. Asare Afrane

    cs.AI · cs.CE · cs.ET · stat.AP · stat.ML

    A consensus anomaly detection framework was applied to monthly malaria surveillance data from Ghana (2014-2023) to identify atypical transmission patterns. Anomalies were highly structured in space and time. Ashanti and Northern Regions accounted for most recurrent anomalies, with persistent hotspots at Tamale, Kumasi, and Accra. A key finding was the spatial distinction between anomaly burden (cumulative cases during anomalous periods) and...

    arxiv.org/abs/2607.21559 · PDF

  2. 02

    Beyond Sycophancy: Structured Resistance and Compliance in LLM Moral Reasoning

    Baihui Wang, Bernard Koch

    cs.AI

    Building socially calibrated large language models, which can learn from others without simply yielding to them, requires more than reducing sycophancy as a one-dimensional failure mode. Models must distinguish when to incorporate others' perspectives from when to maintain a well-grounded moral judgment. We study the broader resistance-compliance process governing this distinction. Across three studies, we show that models' judgment revision...

    arxiv.org/abs/2607.21558 · PDF

  3. 03

    OpenForgeRL: Train Harness-native Agents in Any Environment

    Xiao Yu, Baolin Peng, Ruize Xu, Hao Zou, Qianhui Wu, Hao Cheng, Wenlin Yao, Nikhil Singh, Zhou Yu, Jianfeng Gao

    cs.AI · cs.CL

    Modern AI agents rely on elaborate inference harnesses such as Claude Code, Codex, and OpenClaw to drive multi-turn reasoning, tool use, and access to external systems. While powerful, these complex harnesses also make agents hard to train end-to-end with open infrastructure, whose SFT/RL stacks cannot natively express stateful, multi-process harness inference. To address this, we present OpenForgeRL, an open-source framework for training...

    arxiv.org/abs/2607.21557 · PDF

  4. 04

    MIRROR: Learning from the Other View for Multi-Modal Reasoning

    Wen Ye, Yuxiao Qu, Aviral Kumar, Xuezhe Ma

    cs.AI · cs.LG

    Unlike large language models (LLMs) that exhibit strong reasoning capabilities, vision-language models (VLMs) struggle with visual reasoning, even on geometry problems that admit equivalent text, diagram, and combined diagram+text views. We show that these views often elicit different behaviors: a model may solve a problem from text but fail on the corresponding diagram, or succeed visually while failing textually. This inconsistency suggests...

    arxiv.org/abs/2607.21552 · PDF

  5. 05

    The Boundaries of Automation: A Theory of Persistent Human Participation

    Fares Fourati, Hinrich Schütze, Eyke Hüllermeier, Iryna Gurevych

    cs.AI · cs.CL · cs.ET · cs.LG · cs.MA

    The rapid progress of AI has intensified the long-standing pursuit of automation: replacing human participation with algorithms wherever possible. Implicit in this pursuit is the assumption that humans remain in the loop only because current AI systems are not yet sufficiently capable. This paper challenges that assumption. Rather than asking how far automation can extend, we ask where its conceptual limits lie and argue that human...

    arxiv.org/abs/2607.21547 · PDF

  6. 06

    Same Dangerous Objective, Opposite Advice: Direct Exposure versus Multi-Agent Mediation

    Linjun Li

    cs.AI

    Even a current high-capability LLM can appear safer when shown a dangerous objective directly than when other agents transform and relay its direction. Using OpenAI's gpt-5.6-sol model alias, we test 25 pre-specified mirrored trade-off profiles. Direct exposure to an objective authorizing concealment, fabrication, and pressure produced advice net opposed to its target. After an Id and Censor transformed the same objective into affect and a...

    arxiv.org/abs/2607.21518 · PDF

  7. 07

    Agentic Context Management: Solving Agent Memory and Cost by Treating Them as Lifecycle and Architecture Problems

    Gaurav Dadhich

    cs.AI · cs.IR

    Production AI agents' failures are less often due to an inability to reason well and more often because they cannot manage what is in their reasoning context: conversation histories, large prompts, large tool definitions, and ballooning tool outputs. Agents drown in their own accumulating history while paying a token cost that grows every turn, producing missing recalls within and across conversations. The incumbent response treats this as a...

    arxiv.org/abs/2607.21503 · PDF

  8. 08

    Toward Continuous Assurance for the Democratization of AI Agent Creation in Industry

    Natan Levy, Harel Berger

    cs.AI · cs.ET · cs.MA

    AI agents are increasingly created inside organizations by non-engineering users through low-code, no-code, and conversational development environments. This democratization enables rapid local innovation, but it also creates a reliability gap: agents that appear to users as simple productivity artifacts may depend on changing models, tools, retrieval sources, permissions, prompts, schedules, and external services. These dependencies can...

    arxiv.org/abs/2607.21495 · PDF

  9. 09

    Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks

    Mack Nixon, Liam Wright, Yevgeniya Kovalchuk, Alison Fang-Wei Wu, Martin Danka, Andy Boyd, David Bann

    cs.AI · cs.CL

    Large language models (LLMs) and agents are now widely used tools in code development, with data typically sent to third-party cloud-based models. Their adoption in research using personal data is constrained by governance requirements that typically prohibit data transmission to external services. Locally deployable open-weight models offer an alternative since sensitive data never leave the local environment. We introduce an open-source...

    arxiv.org/abs/2607.21482 · PDF

  10. 10

    AREX: Towards a Recursively Self-Improving Agent for Deep Research

    Shuqi Lu, Chaofan Li, Kun Luo, Zhang Zhang, Hui Wang, Hongwang Xiao, Zheng Liu, Lei Xiong, Jiahao Wang, Sen Wang,...

    cs.AI

    Deep research requires agents to find answers that jointly satisfy multiple constraints. Discovering such answers is costly, whereas verifying a candidate can often be decomposed into tractable constraint-wise checks. This discovery--verification asymmetry suggests that a research agent should do more than simply search longer: it should recursively improve its current answer by verifying intermediate results and using the partially verified...

    arxiv.org/abs/2607.21461 · PDF

  11. 11

    Detecting LLM-Generated Tokens in Human--LLM Coauthored Text

    Yangjun Lu, Hongyi Zhou, Fabian Spill, Kai Ye, Chengchun Shi, Jin Zhu

    cs.AI · stat.ME

    The rise of human-AI collaborative writing has created a growing need for fine-grained detection methods that support localizing likely LLM-generated content in mixed-authorship documents. Existing methods for detecting LLM-generated text mainly focus on document-level classification and cannot identify which parts of the text are generated by LLMs. This paper introduces a new method to address this urgent need. Our method operates at the...

    arxiv.org/abs/2607.21458 · PDF

  12. 12

    Agent-Guided Relational Concept Discovery: Toward Interpretable Surgical Margin Assessment

    Nooshin Maghsoodi, Amoon Jamzad, Robert Policelli, Mohammad Farahmand, Dilakshan Srikanthan, Martin Kaufmann, Kevin...

    cs.AI

    Deep learning models can effectively use Rapid Evaporative Ionization Mass Spectrometry (REIMS) data for surgical margin assessment. However, their clinical adoption remains challenging due to limited generalization to operating room conditions. This difficulty arises because models are typically trained on labeled spectra collected from resected tissue samples, while they must operate on noisy, unlabeled data acquired directly during...

    arxiv.org/abs/2607.21437 · PDF

  13. 13

    Bridging the Gap Between Plausibility and Admissibility: Constraint-Aware Flow Maps for Dynamic Graph Systems

    Michael Romei de Socio, Gian Luca Pozzato, Alessio Merlo

    cs.AI

    Generative models can support decision-making under uncertainty by producing ensembles of plausible future system trajectories, but statistical plausibility does not ensure structural feasibility. This study investigates whether post-sampling symbolic constraints can improve the reliability of generative trajectory modeling in dynamic graph-structured systems. A conditional diffusion model generates future graph-state trajectories from...

    arxiv.org/abs/2607.21421 · PDF

  14. 14

    PATS: Policy-Aware Training Scaffolding for Agentic Reinforcement Learning

    Yipeng Shi, Zhipeng Ma, Yue Wang, Qitai Tan, Yang Li, Peng Chen, Zhengzhou Zhu

    cs.AI

    In long-horizon LLM agent reinforcement learning, weak policies often repeat similar failures, producing uninformative rollout trajectories and limiting effective policy optimization. Existing skill-centric methods improve exploration by optimizing, filtering, or internalizing reusable skills. However, they remain centered on the skills themselves rather than being designed as adaptive training-time support for the evolving policy. To address...

    arxiv.org/abs/2607.21419 · PDF

  15. 15

    Logical Regression for Planning with Axioms

    Connor Little, Christian Muise

    cs.AI

    In automated planning, logical regression is an operation that returns the most general condition necessary for an action to achieve a particular formula. It has many applications, such as allowing for more robust plan execution and providing compact policies for non-deterministic planning. Although relatively simple to calculate in basic planning settings, logical regression becomes significantly more complex when additional factors, such as...

    arxiv.org/abs/2607.21414 · PDF

  16. 16

    Euclid-MCP: A Model Context Protocol Server for Deterministic Logical Reasoning via Prolog

    Bartolomeo Bogliolo

    cs.AI · cs.CL · cs.SE

    Large Language Models (LLMs) excel at natural language understanding and generation but remain unreliable for multi-step logical reasoning, especially in safety-critical or compliance-sensitive domains. Recent neuro-symbolic approaches address this gap by coupling neural models with external symbolic engines, yet most integrations are bespoke and lack a standardized interface for tool-augmented agents. This paper presents Euclid-MCP, an...

    arxiv.org/abs/2607.21412 · PDF

  17. 17

    MSBraM: A Multi-scale Self-supervised Brain Foundation Model for Hierarchical EEG Dynamics Learning

    Tao Zhou, Jing Han, Lingyu Shu, Zixing Zhang

    cs.AI

    Self-supervised foundation models have recently shown strong potential for electroencephalogram (EEG)-based analysis. However, existing approaches struggle to capture the inherently multi-scale temporal structure of EEG signals, where local neural patterns and long-range dependencies jointly encode task-relevant information. This limitation hampers cross-scale representation learning and generalization across diverse downstream tasks. To...

    arxiv.org/abs/2607.21402 · PDF

  18. 18

    Multimodal Pretraining for Generalizable EEG Representation Learning

    Targol Bakhtiarvand, Jugal Kalita, Adham Atyabi

    cs.AI

    Electroencephalography (EEG) models used for epilepsy are often limited to specific datasets and tasks. This limited approach can make it challenging to apply these models across different datasets or in various situations. However, recent studies in foundation models and self-supervised learning suggest that an adaptable EEG backbone could support a range of EEG related tasks. In this study, we have developed a multimodal EEG foundation...

    arxiv.org/abs/2607.21384 · PDF

  19. 19

    Towards Faithful Graph Explanations with Synergistic Edge Effects via Granular Balls

    Jiancu Chen, Shuyin Xia, Guan Wang, Degang Chen, Fan Chen

    cs.AI

    Instance-level explanations aim to reveal the rationale behind a model's decisions for a specific graph. Previous methods explain graph neural networks (GNNs) by selecting important edges to induce subgraphs, where edge importance is assessed by perturbing each edge and observing changes in the model predictions. However, they often neglect the synergistic effects among edges, which are crucial for accurately characterizing edge importance....

    arxiv.org/abs/2607.21381 · PDF

  20. 20

    SPORD: A Simulation-Propose-then-OR-Dispose Approach for Supply Chain Planning

    Jiayin He, Yutong Pan, Sen Yang, Ningxuan Kang, Yongzhi Qi, Jianshen Zhang, Wei Qi, Zuo-Jun Max Shen

    cs.AI

    For years, supply chain planning at e-commerce firms has operated as a collection of isolated projects. Each planning task from static network planning to dynamic warehouse assortment planning requires analysts to spend weeks building models from scratch, calibrating and persuading executives to act on outputs they cannot verify. Three barriers drive this: bespoke models proliferate because standardization is difficult (operational...

    arxiv.org/abs/2607.21354 · PDF

  21. 21

    Regulating autonomous and agentic AI

    Chris Reed, Alex Austria, Anmol Bharuka, Pragnitha Mandava, Khushiya Mujawar, Luka Shakhkulashvili

    cs.AI

    Regulating activities where regulatees use autonomous and agentic AI is challenging. Regulatory assumptions about regulatee knowledge and control no longer hold true; much of that lies elsewhere in the AI supply chain which thus needs to be brought within the scope of regulation. Governance systems for autonomous AI cannot replicate existing governance models, but need a fresh approach. Retrospective supervisory oversight becomes ineffective...

    arxiv.org/abs/2607.21345 · PDF

  22. 22

    Expert Behavior Prior Reinforcement Learning

    Gong Gao, Weidong Zhao, Xianhui Liu, Ning Jia

    cs.AI

    Behavior prior reinforcement learning (BPRL) has emerged as a promising paradigm to improve sample efficiency in online reinforcement learning (RL) by leveraging policy priors derived from offline demonstrations. However, most existing BPRL methods rely on static offline datasets, which often suffer from low data diversity and suboptimal trajectory quality. This reliance restricts the effectiveness of policy priors, hindering both policy...

    arxiv.org/abs/2607.21302 · PDF

  23. 23

    An LLM-Driven Workflow for Automated Process Control Strategy Generation and Tuning from Dynamic Process Models

    Ari Luna Rueda, Eike Cramer, Klaus Hellgardt, Mehmet Mercangöz

    cs.AI · eess.SY

    We present a structured large-language-model-driven workflow for automated multi-variable control design from dynamic process models. The workflow decomposes the design task into constrained code-generation steps: plant-interface construction, normalization, manipulated-variable controlled-variable (MV-CV) pairing, controller specification, closed loop simulation, scenario generation, performance evaluation and Bayesian-optimization (BO)...

    arxiv.org/abs/2607.21292 · PDF

  24. 24

    BasketEvent: Understanding Who Did What and When in Basketball Videos

    Yu Zhang, Jiayuan Rao, Haoning Wu, Weidi Xie

    cs.AI

    Comprehensive basketball video understanding requires resolving not only what event occurs, but also who is responsible and when the key evidence appears. However, exist- ing methods typically treat spatial perception and semantic recognition as isolated tasks, failing to ground events to individual players or pinpoint their temporal boundaries within complex collective dynamics. To bridge this gap, we introduce BasketEvent, a player- centric...

    arxiv.org/abs/2607.21267 · PDF

  25. 25

    Logic Programming Semantics for Causal Processes

    Felix Weitkämper

    cs.AI · cs.LO · cs.PL

    Motivated by challenging modelling issues in the life sciences, we investigate the relationship between logic programming semantics and the eventual states of causal processes compatible with those logic programs. More precisely, we show that while stable models of positive logic programs correspond to the eventual states of processes commencing from a neutral state and continuing undisturbed indefinitely, supported models describe the...

    arxiv.org/abs/2607.21233 · PDF

  26. 26

    ICAE-Bench: Evaluating Coding Agents as Interactive Project Builders

    Zhongyuan Peng, Dan Huang, Chuyu Zhang, Caijun Xu, Changyi Xiao, Shibo Hong, David Lo, Lin Qiu, Xuezhi Cao, Jiyuan...

    cs.AI

    The recent emergence of vibe-coding workflows is changing what coding agents are expected to do. Instead of merely completing code under fully specified instructions, agents are increasingly expected to transform incomplete product intent into working software by combining various abilities including planning, requirement clarification, tool use, debugging, and repository-level construction. Yet existing benchmarks have not fully caught up...

    arxiv.org/abs/2607.21217 · PDF

  27. 27

    How Rules Represent Causal Knowledge: Causal Modeling with Probabilistic Logic Programming

    Kilian Rueckschloss, Felix Weitkaemper

    cs.AI · cs.LO · cs.PL

    Pearl famously argues that causal knowledge enables the prediction of intervention effects. By contrast, purely descriptive knowledge supports only conclusions drawn from observations. His theory of causality, however, is developed exclusively within Bayesian networks and causal models. Consequently, it is largely restricted to acyclic causal relationships, and transferring its ideas to other formalisms risks misinterpretation or...

    arxiv.org/abs/2607.21208 · PDF

  28. 28

    A New Well-Supported Semantics for Description Logic Programs

    Spencer Killen, Jia-Huai You

    cs.AI

    Description logic programs are a powerful formalism for combining rules with ontologies. The well-supported semantics for description logic programs ensures that no answer sets rely on cyclic dependencies. Most popular semantics for logic programming have this property of well-supportedness. We recognize two limitations of the current well-supported semantics for DL programs: its increased computational complexity for the consistency problem...

    arxiv.org/abs/2607.21203 · PDF

  29. 29

    Bound-Founded Semantics for Answer Set Programming with Difference Constraints: Preliminary Report

    Pedro Cabalar, Jorge Fandinno, Nicolas Rühling, Torsten Schaub, Sebastian Schellhorn, Philipp Wanko

    cs.AI · cs.LO

    While the integration of linear constraints has significantly expanded the reach of Answer Set Programming (ASP), existing hybrid solvers often rely on disparate semantic underpinnings that lack a unified logical foundation. We address this gap by introducing a many-sorted variant of the Bound-founded Logic of Here-and-There (HTb), providing a versatile framework capable of characterizing equilibrium models across a wide spectrum of...

    arxiv.org/abs/2607.21201 · PDF

  30. 30

    Identifying Good Rules for Efficient SAT Encodings of Single-Constant Multiplication Using Machine Learning

    Chufeng Jiang, Neng-Fa Zhou

    cs.AI

    The Single Constant Multiplication problem is a fundamental NP-hard optimization task in hardware design, which seeks to decompose a fixed constant using only additions, subtractions, and bit-shifts. Although dynamic programming methods can produce near-optimal SAT encodings for SCM, their encoding cost remains high for large constants. We propose a neuro-symbolic framework that accelerates SCM SAT encoding by identifying good rules for...

    arxiv.org/abs/2607.21188 · PDF

  31. 31

    Differentiable Logic Programming to Mitigate Reasoning Shortcuts in Neurosymbolic Systems

    Akihiro Takemura, Katsumi Inoue

    cs.AI · cs.LO

    Neurosymbolic (NeSy) systems integrate neural networks with logical reasoning to achieve both generalization and interpretability, but recent work has shown they are susceptible to shortcut reasoning behaviors. We propose a novel method using matrix-based differentiable logic programming to mitigate reasoning shortcuts in two phenomena: constraint satisfaction shortcuts, where constraints are satisfied without achieving the intended task, and...

    arxiv.org/abs/2607.21185 · PDF

  32. 32

    Explaining Weather Bulletins via ILP

    Enrico Santi, Alessandro Dal Palù, Agostino Dovier, Talissa Dreossi, Andrea Formisano

    cs.AI · cs.LG · cs.SC

    Inductive Logic Programming (ILP) originated within the Logic Programming community in the Nineties as a framework for combining symbolic learning with declarative knowledge representation. Nowadays, mature ILP frameworks exist and they are capable of learning complex, non-monotonic hypotheses, thus broadening both the modeling capabilities and the scope of real-world applications of ILP. This work is primarily based on the FastLAS2 framework...

    arxiv.org/abs/2607.21184 · PDF

  33. 33

    Safeguards for Speech2Speech LLM-Assistants: A Case Study in Automotive Applications

    Gregor Endler, Sebastian Kraus, Lukas Stappen

    cs.AI

    Recent advances have introduced speech-to-speech (S2S) conversational assistants capable of producing natural-sounding interactions, including non-verbal cues like tonality and mood. In the automotive domain, this enables intuitive and humanlike in-car dialogue experiences. However, integrating these end-to-end assistants limits architectural options for programmable domain-specific safeguards. This paper discusses two implementation...

    arxiv.org/abs/2607.21180 · PDF

  34. 34

    SafeStep: AI-powered Travel Assistance for Elderly People with Frailty or Dementia

    Elderly People with Frailty or Dementia Azul Debenedetti, David Gamez, Franco Such, Nik Kairinos

    cs.AI · cs.MA

    More than a million people in the UK suffer from frailty or dementia, which severely compromise their ability to travel in urban environments. This paper presents SafeStep, an AI-driven travel system that assists elderly users with their journeys. At the core of SafeStep is a novel travel graph representation, which integrates route planning with predictive modelling. For each stage of a journey, the system (i) generates personalized failure...

    arxiv.org/abs/2607.21156 · PDF

  35. 35

    V-DEAL: Diagnosing Video Safety De-Calibration as an Understanding-Refusal Coupling Failure

    Zhetong Zhang, Honghao Fu, Miao Xu, Yiwei Wang, Yujun Cai

    cs.AI

    As Video Large Language Models are increasingly deployed in real-world applications, ensuring their safety alignment has become critical. Counterintuitively, we find that harmful videos paired with benign queries achieve higher attack success rates than the same videos paired with explicitly harmful queries. To understand the underlying mechanism of this vulnerability, we present V-DEAL, a three-level diagnostic framework that jointly...

    arxiv.org/abs/2607.21151 · PDF

  36. 36

    AttriMem: Attribution-Guided Process Feedback for Agent Memory Learning

    Qinfeng Li, Yuntai Bao, Xinyan Yu, Hongze Chen, Wenqi Zhang, Xuhong Zhang

    cs.AI

    Effective memory is crucial for LLM agents, yet constructing it effectively remains challenging. A memory-construction policy decides what information to extract, store, update, compress, or discard as interactions accumulate. Heuristic memory methods rely on subjective, task-specific rules, which can misalign with downstream objectives and limit cross-task adaptability. RL-based methods, by contrast, learn from task feedback but mainly use...

    arxiv.org/abs/2607.21106 · PDF

  37. 37

    Can Generative Recommendation Reach Cold Items? A Temporal Perspective on Semantic-ID Generation

    Jie Peng, Yanping Zheng, Zhewei Zhe, Bin Tong, Guan Wang, Bo Zheng

    cs.AI

    Semantic-ID-based generative recommendation represents items as sequences of shared semantic tokens, enabling token recombination beyond isolated item IDs. However, closed-world recombination does not necessarily imply temporal open-token cold-start induction, where new items enter the item catalog with unseen atomic tokens or weakly supported SID paths. In this work, we revisit SID-based generative recommendation under an absolute-time...

    arxiv.org/abs/2607.21101 · PDF

  38. 38

    Faster IndexTTS-2: Accelerating and Streaming Autoregressive Zero-Shot Text-to-Speech Synthesis on GPUs

    Muyang Du, Shuang Yu, Junjie Lai

    cs.AI

    Autoregressive text-to-speech models achieve strong naturalness but suffer from slow inference due to sequential token generation, limiting their deployment in production applications that require low latency. IndexTTS-2 is a state-of-the-art autoregressive TTS model consisting of a GPT, a flow-matching Diffusion Transformer, and a vocoder. Despite its high synthesis quality, its inference speed barely reaches real-time without streaming or...

    arxiv.org/abs/2607.21042 · PDF

  39. 39

    HiMe: Real-Time Self-Hosted Personal Agent Platform for Health Insights with Wearable Devices

    Wei Liu, Siya Qi, Linhai Zhang, Lorainne Tudor Car, Yulan He

    cs.AI · cs.CL · cs.HC · cs.MA · cs.SE

    Traditional approaches to wearable health signal analysis, such as smartwatches, are constrained by rigid analytical frameworks and limited personalisation. The emergence of LLM agents creates a new opportunity for Personal Health Agentic Analysis, where health insights can be generated adaptively and in context. However, currently there is no open-source locally deployable platform capable of processing personal health data in real time...

    arxiv.org/abs/2607.21019 · PDF

  40. 40

    EmoAgent-R1: Towards Multimodal Emotion Understanding with Reinforcement Learning-based Dynamic Agent Specialization

    Lihuang Fang, Yuchen Zou, kebin Jin, Jinghui Qin

    cs.AI · cs.CV

    Multimodal large language models (MLLMs) have achieved impressive performance in multimodal emotion recognition (MER) tasks and lifted MER to a new level that is complex emotion understanding with advanced video understanding abilities and natural language description. However, existing MLLM-based methods often use a fixed prompt to perceive the emotions, ignoring the dynamicity and complexity of the emotion source in the multimodal inputs....

    arxiv.org/abs/2607.21013 · PDF

  41. 41

    Reexamining zero-shot summarization: Empirical investigation of trustworthiness of LLM-summarizers

    Vasudha Bhatnagar, Purnima Bindal, Vikas Kumar, Raj Kumari Bahl

    cs.AI

    Zero-shot summarization using Large Language Models (LLMs) has significantly advanced the abstractive summarization task by producing coherent and fluent summaries. However, underlying stochasticity of the large language models raises concerns about the stability and trustworthiness of the LLM-generated summaries. This issue has become increasingly important due to proliferation of LLM-generated summaries in educational settings, where...

    arxiv.org/abs/2607.21010 · PDF

  42. 42

    Naju: A Native Discrete State-Space Model with Independent Retention and Writing for Long-Sequence Memory

    Hyuk Lim, Seunghyun Yoon

    cs.AI

    Long-sequence memory tracking places two opposing demands on a recurrent state: near-lossless retention of stored bindings over long horizons, and active overwriting of stale ones. In our diagnostic suite, the strongest efficient baselines tend to solve only one side well. Continuous-time-parameterized state-space models (SSMs) such as Mamba obtain their discrete recurrence by zero-order-hold discretization of a continuous-time system; we...

    arxiv.org/abs/2607.21000 · PDF

  43. 43

    Workflow-Localized Mechanism Learning: Attribution-Guided Repair and Knowledge Reuse for Structured Agent Skills

    Zibin Lin, Shengli Zhang, Taotao Wang, Yihan Xia, Deen Ma, Guofu Liao

    cs.AI

    Agent Skills package reusable procedural knowledge as external artifacts for frozen language-model agents, yet existing optimizers do not jointly resolve where a failure occurs in a workflow, which mechanism caused it, and how relevant knowledge from third-party Skills should be reused locally. We introduce Workflow-Localized Mechanism Learning (WML). Its Node--Mechanism Attribution identifies the failed workflow node, implicated mechanisms,...

    arxiv.org/abs/2607.20999 · PDF

  44. 44

    GuardianAgentBench: Where Agents Fail and How to Guard Them

    Vishal Ishwar Naik, Chenyu Xu, Donna Dong, Hussein Hassan, Abhishek Pradhan, Ofer Mendelevitch, Tallat Shafat, Humayun Irshad

    cs.AI

    As large language model agents increasingly operate autonomously with access to tools and external environments, ensuring their safe and reliable behavior becomes critical. We present GuardianAgentBench (GABench), a benchmark of 580 scenarios across six domains evaluated on three production-ready frameworks: LangChain, LlamaIndex, and Vectara. The benchmark incorporates rigorous multi-stage validation and five adversarial attack modes....

    arxiv.org/abs/2607.20982 · PDF

  45. 45

    Beyond Independent Optimization: Compression, MoE Routing, and Quantization Interactions in Multimodal Edge Intelligence

    Jay Gor, Karm Dave, Akshita Abrol, Rajesh Gupta, Sudeep Tanwar, Zhengkui Wang

    cs.AI

    Efficient multimodal inference is increasingly constrained not only by model quality or FLOP count, but also by the cost of preserving, moving, routing, caching, and quantizing multimodal representations under latency, memory, and energy constraints. This paper reviews recent advances in efficient vision-language and multimodal large language models, covering visual token compression, video token management, KV-cache optimization,...

    arxiv.org/abs/2607.20981 · PDF

  46. 46

    Delivery, Not Storage: Cue-Anchored Working Memory as a Harness Property for Coding Agents

    Swapnanil Saha

    cs.AI · cs.SE

    Coding agents ship with one kind of memory: documents. Instruction files, plan artifacts, and auto-written memory directories are deliberately authored and deliberately retrieved: the agent must choose to write them and choose to read them back. Human expertise runs on a second tier that never gets written down: situationally-bound operational facts (gotchas, locations, local conventions) encoded as a side effect of the work and retrieved...

    arxiv.org/abs/2607.20972 · PDF

  47. 47

    From Scalars to Time Series: Rethinking Implicit Neural Representations for Time-Varying Volumetric Data

    Weihan Zhang, Xuan Zhao, Yenwen Peng, Yuqi Chen, Jun Tao

    cs.AI

    Implicit neural representations (INRs) for time-varying volumetric data are typically trained using dense sampling over spatiotemporal coordinates, where each observation corresponds to a single point in space and time. This coordinate-wise formulation requires extensive sampling during optimization, leading to high computational cost and inefficient use of temporal structure. In this work, we revisit this design choice and show that dense...

    arxiv.org/abs/2607.20970 · PDF

  48. 48

    Clustered Edge Intelligence: Beyond Just Convergence of Edge Computing and AI

    Chinmaya Kumar Dehury, Boris Sedlak, Alaa Saleh, Ilir Murturi, Lauri Loven, Satish Narayana Srirama, Praveen Kumar Donta

    cs.AI · cs.DC

    We are moving from an information age to the age of intelligence. A decade, or possibly less than that, data will not be the gold anymore rather the derived intelligence out of the data and the information we posses from the edge of the network. Existing Edge Intelligence research focuses mainly on two directions: using AI for edge resource management and deploying lightweight AI models on edge devices. However, existing edge computing...

    arxiv.org/abs/2607.20937 · PDF

  49. 49

    SciExplore: Evaluating Autonomous Agents from Scientific Navigation to Information Integration

    Yinhao Tang, Youqing Fang, Yanan Sun, Wenran Liu, Weiming Zhang, Bin Liu, Kuikun Liu, Wenwei Zhang, Kai Chen

    cs.AI

    Scientific research involves complex information-seeking and reasoning workflows across heterogeneous sources. However, existing benchmarks primarily emphasize general-domain retrieval or static scientific question answering, and therefore fail to assess key capabilities required in realistic scientific research workflows. We introduce SciExplore, a benchmark designed to evaluate scientific information-seeking and reasoning capabilities of...

    arxiv.org/abs/2607.20926 · PDF

  50. 50

    Representing Entity Importance in AI Knowledge Systems: A Dual-Signal Framework of Audience Evaluation and Structural Authority

    Shen Xu

    cs.AI

    AI knowledge systems require representations of entity importance for retrieval, recommendation, evidence selection, and knowledge-intensive reasoning. Yet importance is often reduced to a single score derived from either human response or graph structure. Such compression may discard distinctions that matter when an AI system must choose among entities for different tasks. This study introduces an interpretable dual-signal representation in...

    arxiv.org/abs/2607.20925 · PDF

  51. 51

    OPOD: On-Policy Omni Distillation

    Tong Zhao, Yuyang Hu, Reed Li, Yu Lu, Haibo Shi, Yutao Zhu, Zhicheng Dou

    cs.AI

    Omni-modal models can handle text, images, and audio in one system, but improving all of these abilities together remains difficult. Training a single model on pooled multimodal data often fails to match models specialized for individual modalities. On-policy distillation (OPD) offers a way to combine such specialists: the student generates a response, and a teacher evaluates that same response, so the student learns directly from behaviors...

    arxiv.org/abs/2607.20918 · PDF

  52. 52

    Traceable Scholarship: Page Anchors and Ariadne's Thread for Humanistic Inquiry in the Age of Generative AI

    Deyu Jing

    cs.AI · cs.DL

    Generative AI lets large language models produce scholarly-looking text within seconds, yet fluency does not equal valid explanation. The deepest risk is not factual error alone but the appearance that an explanation is already established without clear sources, page numbers, editions, or evidence. We liken the page anchor to Ariadne's thread: within the labyrinth of generative fluency, it is the thread that leads the scholar back to the...

    arxiv.org/abs/2607.20916 · PDF

  53. 53

    Source-Prior-Driven Selective Adaptation for Efficient Diffusion Model Finetuning

    Yi Xiong, Yuan-Yuan Cheng, Xiao-Ming Fu

    cs.AI

    Fine-tuning large diffusion models for new domains or styles involves a trade-off: improving target-specific generation often degrades the pretrained model's broad generative capability. Existing full and parameter-efficient fine-tuning methods typically handle this trade-off only implicitly. In this work, we propose a novel source-prior-driven selective adaptation method to efficiently fine-tune diffusion models, achieving a favorable...

    arxiv.org/abs/2607.20913 · PDF

  54. 54

    Is Deep Research Reliable? Misleading Knowledge Induces False Conclusions

    Pengyu Zhu, Lijun Li, Longju Yang, Sen Su

    cs.AI

    Deep Research agents extend LLM-based assistants into long-horizon workflows involving planning, retrieval, evidence synthesis, and report generation, yet their reliability in open information environments remains underexplored. A key concern is whether apparently credible but factually misleading knowledge encountered in such environments can propagate through these workflows and be adopted as false conclusions in final reports. To study...

    arxiv.org/abs/2607.20891 · PDF

  55. 55

    Code Monitor Red Teaming for Public-Test-Passing Code

    Junchi Liao, Jiawen Deng, Fuji Ren

    cs.AI

    Visible tests are a common gate for LLM-generated code, but passing them does not certify specification correctness. We study a deployment-like monitoring problem: after code has passed public tests, can a weaker LLM verifier identify the residual hidden bugs? We introduce Code Monitor Red Teaming, a monitor-red-teaming protocol that fixes a public-check information boundary while varying generator pressure, verifier scaffolding, and...

    arxiv.org/abs/2607.20852 · PDF

  56. 56

    Auditing Evidence Use in Medical LLM Diagnosis

    Junchi Liao, Jiawen Deng, Fuji Ren

    cs.AI

    Medical LLMs are often evaluated by whether they select the correct diagnosis, but diagnostic accuracy alone does not show whether the model used the case evidence appropriately. We present a behavioral audit of evidence use in medical diagnosis. For each case, we decompose patient information into evidence units, score candidate diagnoses under controlled evidence subsets, and mine low-order interactions in diagnostic margins. Because...

    arxiv.org/abs/2607.20848 · PDF

  57. 57

    Auditing Provenance Sensitivity in LLM Agent Action Selection

    Junchi Liao

    cs.AI

    LLM agents choose tools and arguments from context that mixes user requests, tool outputs, retrieved records, memory, and untrusted text. Evidence can be relevant without being authorized to determine a decision, so a correct action need not be grounded only in permitted evidence. We introduce a target-specific authorization audit that labels context factors separately for each tool and argument target. Its primary test holds the task,...

    arxiv.org/abs/2607.20827 · PDF

  58. 58

    Efficient and Interpretable Body-Based Emotion Recognition with Lightweight Temporal Convolutional Networks

    Christian Arzate Cruz, Stefanos Gkikas, Houshyar Asadi

    cs.AI

    Body-based emotion recognition is important for real-time affective systems, but graph-based skeleton models can be computationally expensive. This paper studies whether lightweight temporal convolutional networks (TCNs) can provide an efficient and interpretable alternative for body-based emotion classification. We evaluate a family of TCN models on DIEM-A and compare them with a graph-based time-series graph (G-TSG) baseline using accuracy,...

    arxiv.org/abs/2607.20820 · PDF

  59. 59

    Enhancing Explainable Cardiac Diagnosis with Guide-Grounded Multimodal LLMs

    Hai-Nam Duy Vuong, Duy-Anh Bui, Trong-Nghia Nguyen, Kim-Ngan Thi Nguyen, Trang Mai Xuan, Tien-Cuong Nguyen, Van-Dem...

    cs.AI

    The electrocardiogram (ECG) is a cornerstone of cardiac as- sessment, yet clinical deployment of deep learning models remains con- strained by limited interpretability and the hallucination risk of large language models (LLMs). Existing CNN+Grad-CAM+multimodal LLM frameworks can generate ECG reports, but their explanations are often only weakly grounded in established diagnostic criteria, reducing trust- worthiness and reproducibility. We...

    arxiv.org/abs/2607.20814 · PDF

  60. 60

    Profiling Lightweight Large Language Models

    Tomohiro Harada, Enrique Alba, Gabriel Luque

    cs.AI

    Lightweight large language models (LLMs) are increasingly being deployed locally on personal computers and are expected to play a growing role in resource-constrained edge and mobile environments. In such settings, energy consumption, execution time, and memory usage directly affect practical usability, yet existing evaluations of LLM efficiency largely rely on proxy descriptors such as parameter count or FLOPs, often decoupled from task...

    arxiv.org/abs/2607.20806 · PDF

  61. 61

    Can an AI System Be Creative? A Critical Perspective from Art and Engineering

    Ivan Magrin-Chagnolleau

    cs.AI · cs.HC · cs.MM

    This paper examines the question of whether artificial intelligence (AI) systems can be creative, approached from the dual perspective of a researcher trained in electrical engineering, pattern recognition, machine learning, and neural networks, who has also spent most of his life engaged in the arts as actor, stage and film director, writer, composer, and visual artist, and in philosophy. Drawing on Margaret Boden's foundational framework,...

    arxiv.org/abs/2607.20796 · PDF

  62. 62

    Refusal-Gated Decoding: Preserving Refusal Behavior Under High-Temperature Sampling

    Phillip Howard, Xin Su, Allen Roush, Manikandan Ravikiran, Amir Abdullah

    cs.AI · cs.CL

    High-temperature sampling is one of the primary mechanisms for increasing diversity in LLMs. Recent advances in truncation-based sampling techniques have helped mitigate drawbacks of high-temperature sampling such as neural text degeneration, thereby enabling greater diversity in LLM outputs without sacrificing coherence. However, increasing the entropy of the token probability distribution via high temperatures has also been shown to weaken...

    arxiv.org/abs/2607.20791 · PDF

  63. 63

    The Human-AI Substitution Principle: When will you be replaced by AI in your organization?

    Bonny Banerjee, Shreya Singh

    cs.AI · econ.GN

    Artificial Intelligence (AI) is rapidly transforming organizations, raising a fundamental organizational and economic question: when will a human employee be replaced by AI? We present an analytical model for studying Human--AI Task Allocation (HAT) in hierarchical organizations. A central feature of the HAT model is that it formally encodes the economic asymmetry between human skill acquisition and AI capability scaling. The HAT model allows...

    arxiv.org/abs/2607.20781 · PDF

  64. 64

    ArbiGraph: Arbitrarily Scalable Verifiable Task Graphs for Evaluating Context Management

    Pavel Golikov, Evgenii Opryshko, Gennady Pekhimenko, Mark C. Jeffrey

    cs.AI

    We introduce ARBIGRAPH, a benchmark generator for evaluating whether tool-assisted language agents can retain, update, compose, and discard task-relevant context across extended reasoning workflows. ARBIGRAPH represents each task as a natural-language problem with an executable Python solver, and composes tasks through typed intermediate states, instantiated here as scalar and list values. This design enables controllable task graphs whose...

    arxiv.org/abs/2607.20764 · 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.