cs.AI · 2026-06-24 · No. 33

Artificial Intelligence, 2026-06-24.

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

01 — The papers

66 entries
  1. 01

    OpenThoughts-Agent: Data Recipes for Agentic Models

    Negin Raoof, Richard Zhuang, Marianna Nezhurina, Etash Guha, Atula Tejaswi, Ryan Marten, Charlie F. Ruan, Tyler...

    cs.AI

    Agentic language models dramatically expand the applications of AI yet little is publicly known about how to curate training data for broadly capable agents. Existing open efforts such as SWE-Smith, SERA, and Nemotron-Terminal typically target a single benchmark, leaving open the question of how to train models that generalize across diverse agentic tasks. The OpenThoughts-Agent (OT-Agent) project addresses this gap with a fully open data...

    arxiv.org/abs/2606.24855 · PDF

  2. 02

    World Models in Pieces: Structural Certification for General Agents

    Yikai Lu, Yifei Wu, Xinyu Lu, Tongxin Li

    cs.AI

    In the big-world regime, agents cannot be universally capable and their ability is inevitably specialized across a world model in pieces. Consequently, standard uniform guarantees fail to distinguish between the understanding of critical bottlenecks and irrelevant failures. We first formalize this limitation by proving that general agents are not universal, rendering standard worst-case analysis uninformative. To overcome this, we introduce...

    arxiv.org/abs/2606.24842 · PDF

  3. 03

    Matching Tasks to Objectives: Fine-Tuning and Prompt-Tuning Strategies for Encoder-Decoder Pre-trained Language Models

    Ahmad Pouramini, Hesham Faili

    cs.AI · cs.CL

    Prompt-based learning has emerged as a dominant paradigm in natural language processing. This study explores the impact of diverse pre-training objectives on the performance of encoder-decoder pre-trained language models across generation and question answering tasks, with a focus on commonsense knowledge retrieval and completion. We highlight the benefits of incorporating multiple objectives during both pre-training and fine-tuning stages....

    arxiv.org/abs/2606.24841 · PDF

  4. 04

    Grading the Grader: Lessons from Evaluating an Agentic Data Analysis System

    Tian Zheng, Kai-Tai Hsu

    cs.AI · stat.AP

    Agentic data analysis systems produce rich outputs, including code, numerical results, and verbal diagnostics. This makes them more challenging to evaluate than single-turn LLM responses. It is therefore necessary to distinguish genuine disagreement between an agent's output and a ground-truth answer from grading artifacts. We investigate how reliably automated graders assess such a system and what strategies improve grading quality by...

    arxiv.org/abs/2606.24839 · PDF

  5. 05

    Accuracy and Satisfaction in Multi-Turn LLM Dialogues for NFR Assessment

    Ali Pourghasemi Fatideh, Wilder Baldwin, Maria Dhakal, Collin McMillan, Sepideh Ghanavati

    cs.AI

    LLM-based dialogue assistants have become mainstream tools for software developers, yet current evaluation benchmarks focus exclusively on functional correctness. This leaves a critical gap in assessing the quality and accuracy of these conversations when handling Non-Functional Requirements (NFRs), which are inherently vague, context-dependent, and involve many parts of a program. Evaluating how well these systems support collaborative...

    arxiv.org/abs/2606.24834 · PDF

  6. 06

    Difference-Making without Making a Difference

    Sander Beckers

    cs.AI

    Over a series of seven papers, Andreas & Günther have introduced seven definitions of actual causation and have classified them as belonging to three different, competing, types of accounts: factual difference-making, counterfactual difference-making, and regularity-based. I show that their most recent - factual difference-making - definition instantiates all three types, thereby proving that these are distinctions without a difference. I...

    arxiv.org/abs/2606.24832 · PDF

  7. 07

    Solving Inverse Problems of Chaotic Systems with Bidirectional Conditional Flow Matching

    Peiyan Hu, Jian Zhang, Jiashu Pan, Ruiqi Feng, Tao Zhang, Zhi-Ming Ma, Yuan-Sen Ting, Gongjie Li, Tailin Wu

    cs.AI

    Modeling chaotic systems is crucial yet challenging. Inverse problems in chaotic dynamics, namely inferring initial conditions from final states, remain largely unsolved because of ill-posedness, non-uniqueness, instability, and potentially chaotic time-reverse dynamics. We address this open problem with Bidirectional Conditional Flow Matching (Bi-CFM), which learns bidirectional mappings between distributions of initial and final states to...

    arxiv.org/abs/2606.24824 · PDF

  8. 08

    Assessing Distribution Shift in Human Activity Recognition for Domain Generalization

    Rebecca Adaimi, Edison Thomaz

    cs.AI · cs.HC

    While the field of Human Activity Recognition (HAR) continues to draw interest from researchers and advance in important ways, some key challenges remain. One of the most difficult aspects of building HAR models that show good performance in real-world settings is dealing with data diversity from device and sensor heterogeneity, and contextual changes that are intrinsic to real-world applications. While data diversity in HAR has been...

    arxiv.org/abs/2606.24781 · PDF

  9. 09

    BluTrain: A C++/CUDA Framework for AI Systems

    Adhitya Charan, Adwaid Suresh, Anuj Kumar, Aparna A, Dhanakumar K, Dharun M S, Dinesh G, Goutham Kumar Reddy K,...

    cs.AI

    Progress in deep learning is, at scale, more a matter of systems engineering than of modelling: the behaviour of a model in training (its throughput, its memory footprint, and the numerical fidelity of the result) is determined less by the architecture itself than by how that architecture is expressed on the hardware. To achieve absolute control over this hardware expression while abstracting away systems complexity to make modelling seamless...

    arxiv.org/abs/2606.24780 · PDF

  10. 10

    Can Scale Save Us From Plasticity Loss in Large Language Models?

    J. Fernando Hernandez-Garcia, Tomás Figliolia, Beren Millidge

    cs.AI

    The loss of plasticity - the ability of a network to learn new information after having already learned older information - is a fundamental challenge in creating artificial neural networks capable of continual learning. Although this phenomenon has been known for decades, it has mostly been studied in older, relatively small architectures and rarely in natural-language domains. To determine whether loss of plasticity remains a problem in the...

    arxiv.org/abs/2606.24752 · PDF

  11. 11

    Scaling Laws for Task-Specific LLM Distillation

    Lavinia Ghita, Dhruv Desai, Ioana Boier

    cs.AI · cs.CE

    Large Language Models (LLMs) achieve strong performance across a growing range of domains, yet their scale poses deployment challenges in applications where latency and cost constraints are critical. This paper derives empirical scaling laws for domain-specific LLM compression, quantifying how in-domain and general knowledge performance scale with dataset size, compression ratio, supervision format, and iterative pruning schedule. Using...

    arxiv.org/abs/2606.24747 · PDF

  12. 12

    Decentralised AI Training and Inference with BlockTrain

    Peter Toth

    cs.AI

    Frontier AI training is increasingly shaped by access to dense, centrally controlled accelerator clusters. This creates a structural advantage for hyperscalers and large centralized laboratories, and makes open or independent AI efforts depend on scarce capital, privileged infrastructure, and data-center geography. We present Spheroid BlockTrain, a decentralized training protocol in which a model is partitioned into independently trainable...

    arxiv.org/abs/2606.24722 · PDF

  13. 13

    Cost-Optimal Decision Diagrams for Stochastic Boolean Function Evaluation

    Xia Zong, Tuomo Lehtonen, Jussi Rintanen

    cs.AI

    In many decision-making scenarios, acquiring information incurs different costs. We consider the problem of constructing a deterministic evaluation strategy that minimizes the expected cost of evaluating a propositional formula under variable costs and a probability distribution over truth assignments. We present a branch-and-bound algorithm with variable-selection heuristics, pruning, and caching. To the best of our knowledge, it is the...

    arxiv.org/abs/2606.24672 · PDF

  14. 14

    LaGO: Latent Action Guidance for Online Reinforcement Learning

    Kuan-Yen Liu, Ren-Jyun Huang, Ti-Rong Wu

    cs.AI

    Large language models (LLMs) have shown strong potential for planning and sequential decision-making, but prior work often relies on using them as direct controllers, which requires precise action generation and can be unreliable in practice. This paper proposes Latent Action Guidance for Online Reinforcement Learning (LaGO), a framework that uses a pretrained LLM as a latent action prior to softly guide online policy optimization, rather...

    arxiv.org/abs/2606.24669 · PDF

  15. 15

    CineCap: Structured Reasoning with Spatio-Temporal Anchors for Cinematographic Video Captioning

    Xinyu Mao, Yuhui Zeng, Xiaokun Liu, Wenyu Qin, Meng Wang, Xin Tao, Pengfei Wan, Xiaohan Xing, Max Meng

    cs.AI

    Cinematographic captioning aims to describe how a video is filmed using professional film-language concepts such as camera movement, shot size, depth of field, composition, and shooting angle. This capability is important for fine-grained video understanding and controllable movie-quality video generation, yet remains underexplored in existing multimodal large language models. Unlike question-answering-based evaluation of cinematic...

    arxiv.org/abs/2606.24636 · PDF

  16. 16

    SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation

    Chenyang Zhu, Jiayu Yao, Kushal Chawla, Youbing Yin, Nathan Wolfe, Pengshan Cai, Jingyu Wu, Spencer Hong, Sangwoo...

    cs.AI

    As autonomous agents tackle increasingly complex multi-step, multi-agent tasks, their execution trajectories have scaled beyond the constraints of even the largest context windows. Current methods for effectively diagnosing agent failures load the full trajectory into an LLM's context window, which suffers from attention dilution and fails when agentic traces inevitably exceed context limits. To address this, we introduce SAFARI (Scaling...

    arxiv.org/abs/2606.24626 · PDF

  17. 17

    Themis: An explainable AI-enabled framework for Reinforcement Learning with Human Feedback

    Andreas Chouliaras, Luke Connolly, Dimitris Chatzpoulos

    cs.AI · cs.HC

    Training safe Reinforcement Learning (RL) systems is inherently challenging, with no guarantee of avoiding unwanted behaviors. The most effective defenses against this are (i) transparency through explainability and (ii) alignment via human feedback. While both show promising results, no publicly available framework currently combines them. To address this, we introduce Themis, an XAI-enabled testing and evaluation framework for Reinforcement...

    arxiv.org/abs/2606.24622 · PDF

  18. 18

    When CQs Go Wrong: Challenges in CQ Verification with OE-Assist

    Anna Sofia Lippolis, Mohammad Javad Saeedizade, Robin Keskisärkkä, Aldo Gangemi, Eva Blomqvist, Andrea Giovanni Nuzzolese

    cs.AI

    Competency Questions (CQs) are the central component of CQ-verification, an established process in which an ontology is evaluated against a set of natural language questions to determine whether the intended purpose of the ontology has been properly modelled. However, CQ-verification is often time-consuming and error-prone, as it requires careful interpretation of linguistic nuances and precise alignment with formal ontology constructs....

    arxiv.org/abs/2606.24619 · PDF

  19. 19

    Abstractions of Queries in Ontology-Based Data Access

    Michel Leclère, Marie-Laure Mugnier, Guillaume Pérution-Kihli

    cs.AI · cs.DB

    In ontology-based data access (OBDA), multiple data sources are integrated via mappings to an ontology. We consider an OBDA setting based on existential rules and the certain answer semantics. We address the recent issue of query abstraction, which consists of abstracting data queries by translating them to the ontology layer. Since a perfect abstraction may not exist, the notions of minimally complete and maximally sound abstractions have...

    arxiv.org/abs/2606.24618 · PDF

  20. 20

    ScaleToT: Generalizing Structured LLM Reasoning for Billion-Scale Low-Activity User Modeling

    Tianbao Ma, Chang Xi, Yichuan Zou, Chengen Li, Linxun Chen, Zilong Lu, Yanan Niu, Zhaojie Liu, Han Li, Kun Gai

    cs.AI

    Accurate user modeling often depends on rich interaction histories, which are unavailable for billions of low-activity users. Large Language Models (LLMs) can infer latent user states from static profiles, but this reasoning becomes unreliable when profiles are sparse, and applying an LLM to billions of users is prohibitively expensive. We present ScaleToT, which learns structured reasoning from a small LLM-processed subset and extends it to...

    arxiv.org/abs/2606.24605 · PDF

  21. 21

    Uncertainty-Aware Longitudinal Forecasting of Alzheimer's Disease Progression Using Deep Learning

    Arya Hariharan, Shreyank N Gowda, Anala M R

    cs.AI

    Longitudinal modelling of Alzheimer's disease progression is clinically useful only if it can describe not just the most likely next diagnosis, but how a patient may evolve over time and how reliable that forecast is. Most deep learning approaches reduce this problem to single-step classification, treating cognitively normal, mild cognitive impairment, and dementia as flat categories while providing limited insight into how uncertainty...

    arxiv.org/abs/2606.24604 · PDF

  22. 22

    ASALT: Adaptive State Alignment for Lateral Transfer in Multi-agent Reinforcement Learning

    Anurag Akula, Satheesh K. Perepu, Abhishek Sarkar, Kaushik Dey

    cs.AI · cs.LG

    Multi-agent reinforcement learning (MARL) addresses the problem of training multiple agents that pursue collaborative, competitive, or mixed objectives. Prior work has investigated transfer learning between source and target domains in MARL; however, the majority of existing approaches impose the constraint that the dimensionalities of the observation space and the global state space must be identical across domains. In this paper, we...

    arxiv.org/abs/2606.24601 · PDF

  23. 23

    AdversaBench: Automated LLM Red-Teaming with Multi-Judge Confirmation and Cross-Model Transferability

    Khanak Khandelwal

    cs.AI · cs.CL

    Scaling adversarial evaluation of large language models requires both a method for generating hard inputs and a reliable way to confirm that resulting failures are real. We present AdversaBench, an end-to-end red-teaming pipeline that mutates seed prompts with five structured operators, queries a target model, and confirms failures through a three-judge panel with a meta-judge tiebreaker. We report experiments on 45 seeds across three...

    arxiv.org/abs/2606.24589 · PDF

  24. 24

    LLMs Prompted for Legal Context Object More: Overrefusal from Small On-Premises LLMs in Criminal Legal Context

    Anastasiia Kucherenko, François Brouchoud, Dimitri Percia David, Andrei Kucharavy

    cs.AI

    While the validity of LLMs' use in the legal context remains subject to ethical and legal debate, legal professionals are already experimenting with personal LLMs, if only for translation and reformulation. However, even such a seemingly innocuous use can introduce biases through case processing speed if LLM assistants selectively refuse assistance on certain topics. To better anticipate such biases, we investigate several modern small LLMs...

    arxiv.org/abs/2606.24585 · PDF

  25. 25

    Quant Convergence: Bridging Classical Value Investing and Modern Factor Models for Systematic Equity Selection

    Augusto Eiji Yamazaki, Hugo Garrido-Lestache Belinchon

    cs.AI

    Modern finance relies heavily on complex machine learning models to find patterns in the stock market. However, as these AI models get more complicated, they often memorize short-term market noise instead of finding companies with real, lasting value. We designed this research to test if Benjamin Graham's classic value investing rules could act as a mathematical "low-pass filter" to keep these modern models in check. We built three different...

    arxiv.org/abs/2606.24575 · PDF

  26. 26

    Governed Shared Memory for Multi-Agent LLM Systems

    Yanki Margalit, Nurit Cohen-Inger, Erni Avram, Ran Taig, Oded Margalit

    cs.AI

    Multi-agent LLM environments require robust mechanisms for shared knowledge management. This paper formalizes the fleet-memory problem and identifies four foundational failure modes: unauthorized leakage, stale propagation, contradiction persistence, and provenance collapse. To address these, we define explicit systems-level primitives: scoped retrieval, temporal supersession, provenance tracking, and policy-governed memory propagation. These...

    arxiv.org/abs/2606.24535 · PDF

  27. 27

    Reinforcement Learning for Computer-Use Agents with Autonomous Evaluation

    Marta Sumyk, Oleksandr Kosovan

    cs.AI · cs.HC

    Computer-Use Agents (CUAs) execute high-level user goals by perceiving and acting directly within graphical user interfaces. However, reinforcement learning for CUAs remains difficult because open-ended desktop environments rarely provide scalable, machine-readable reward signals: task success is often visually grounded and hard to specify with handcrafted reward functions or dense manual labels. We propose an RL fine-tuning framework that...

    arxiv.org/abs/2606.24515 · PDF

  28. 28

    A specialized reasoning large language model for accelerating rare disease diagnosis: a randomized AI physician assistance trial

    Haichao Chen, Songchi Zhou, Zhengyun Zhao, Shikai Hu, Xianghong Jin, Hongwei Ji, Li He, Shuli Li, Yiming Qin, Xin...

    cs.AI · cs.CL

    Rare diseases affect millions of individuals worldwide, yet timely diagnosis remains a major public health challenge due to scarcity of specialized clinical expertise. While large language models (LLMs) show promise to support rare disease diagnosis, current models are constrained by insufficient clinical deployability, limited clinically grounded evidence, and scarcity of training data. Here we present RaDaR (Rare Disease navigatoR), an...

    arxiv.org/abs/2606.24510 · PDF

  29. 29

    On the Smallness of the Large Language Models Scaling Exponents

    Sauro Succi, Peter V. Coveney, Alex Hansen

    cs.AI

    We discuss reasons why the scaling exponents of current Large Language Models (LLMs) applications are indicating an unsustainable regime in terms of energy resources. We further show that attributing the smallness of such exponents to a numerical bias due to the neglect of a non-zero value of the loss function in the limit of infinite data (``pedestal effect") does not remove the unsustainability issue. Finally, the effects of the smoothness...

    arxiv.org/abs/2606.24504 · PDF

  30. 30

    The Latent Bridge: A Continuous Slow-Fast Channel for Real-Time Game Agents

    Bojie Li, Noah Shi

    cs.AI

    A real-time agent for general computer use - with games as the most demanding case - must act within tens of milliseconds while still planning over seconds. These two regimes sit at opposite ends of the latency-quality tradeoff. A reasoning VLM (Qwen3-VL-8B-Thinking) deliberates effectively but requires ~1.5 s per response - far too slow for a 15 Hz control loop. In contrast, a reactive VLM (MiniCPM-o 4.5) acts in milliseconds but...

    arxiv.org/abs/2606.24470 · PDF

  31. 31

    CompressKV: Semantic-Retrieval-Guided KV-Cache Compression for Resource-Efficient Long-Context LLM Inference

    Xiaolin Lin, Jingcun Wang, Olga Kondrateva, Yiyu Shi, Bing Li, Grace Li Zhang

    cs.AI

    Long-context large language model (LLM) inference is increasingly constrained by the memory footprint and decoding cost of key-value (KV) caches, limiting sustainable deployment on resource-constrained hardware. Existing KV cache eviction methods typically apply heuristic token scoring over all heads in GQA-based LLMs. These methods ignore the different functionalities of attention heads, leading to the eviction of critical tokens and thus...

    arxiv.org/abs/2606.24467 · PDF

  32. 32

    Bayesian control for coding agents

    Theodore Papamarkou, Vladislav Smirnov, Viktor Mazanov, Artem Vazhentsev, Preslav Nakov, Timothy Baldwin, Artem Shelmanov

    cs.AI · cs.CL

    Modern coding agents pair LLM generators with various tools, including cheap diagnostics and expensive verifiers. The tool-use decisions are typically governed by orchestrators that often use fixed rules and ignore uncertainty. We formulate orchestration as cost-sensitive sequential hypothesis testing: a Bayesian controller maintains a belief over candidate correctness and dynamically decides whether to gather more evidence, refine the...

    arxiv.org/abs/2606.24453 · PDF

  33. 33

    ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling

    Heng Ping, Arijit Bhattacharjee, Peiyu Zhang, Shixuan Li, Wei Yang, Ali Jannesari, Nesreen Ahmed, Paul Bogdan

    cs.AI

    Mixture-of-Agents (MoA) architectures improve inference-time scaling by organizing multiple LLM agents into layered reasoning pipelines. However, existing MoA variants fail to sustain gains as depth increases, exhibiting degradation, early plateauing, or saturation. We propose ReM-MoA, a memory-augmented MoA framework that sustains scaling through two mechanisms: (1) a Ranked Reasoning Memory that persistently stores and ranks reasoning...

    arxiv.org/abs/2606.24437 · PDF

  34. 34

    Can Aggregate Invariants Accelerate Continuous Subgraph Matching? Limits, Laws, and a Dynamic Spectral Index

    Minghao Chen, Jiale Zheng

    cs.AI · cs.DB · cs.DS

    Spectral filtering recently delivered substantial pruning for \emph{static} subgraph matching: Laplacian interlacing rejects candidates whose neighborhoods cannot host the query. We study whether such aggregate structural tests can accelerate \emph{continuous} subgraph matching (CSM) over dynamic graphs, and answer in three parts. First, lazily maintained spectral bounds are infeasible exactly where spectral pruning has value: we characterize...

    arxiv.org/abs/2606.24421 · PDF

  35. 35

    Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems

    Bingnan Xiao, Chenhao Yang, Wei Ni, Xin Wang, Tony Q. S. Quek

    cs.AI

    Network operators' changing policies, service requirements, and stringent real-time constraints render existing methods designed with fixed objectives and constraints ineffective. This paper presents Agentic long-term performance optimization (Agentic-LTPO), a nested bilevel optimization framework that can be applied to adaptive physical layer problem configuration. The key idea is to employ agentic AI to generate upper-level configurations...

    arxiv.org/abs/2606.24416 · PDF

  36. 36

    Cycle-Consistent Neural Explanation of Formal Verification Certificates

    Andoni Rodriguez, Alberto Pozanco, Daniel Borrajo

    cs.AI

    Formal verification produces machine-checkable certificates that attest to the satisfaction or violation of temporal properties, yet these certificates remain opaque to non-specialist stakeholders. We propose a cycle-consistent neural architecture that generates faithful natural language explanations of verification certificates. A forward network NN1 maps certificates to explanations, and an inverse network NN2 reconstructs certificates from...

    arxiv.org/abs/2606.24414 · PDF

  37. 37

    ATRIA: Adaptive Traceable ECG Reporting with Iterative Agents

    Donggyun Hong, Kyuhwan Lee, Junmyung Kwon, Yong-Yeon Jo

    cs.AI

    Existing ECG report generation is tightly coupled -- interpretation and reporting fused end-to-end, so errors propagate without stage-level recourse -- while agent-based systems decouple tasks but remain single-pass, never revisiting earlier outputs. Clinical ECG reporting instead unfolds iteratively, requiring progressive context integration and bidirectional editing. We present \textsc{ATRIA}, a multi-agent ECG reporting system that mirrors...

    arxiv.org/abs/2606.24392 · PDF

  38. 38

    Age of LLM: A Strategic 1v1 Benchmark for Reasoning, Diplomacy and Reliability of Large Language Models under Fog of War

    Arnaud Ricci

    cs.AI · cs.CL · cs.GT · cs.MA

    We introduce Age of LLM, a turn-based 1v1 benchmark in which two LLMs face off on a 13x7 grid to destroy the enemy base. Three stressors are deliberate: fog of war, full diplomacy (messages, ceasefires, ultimatums; uranium kept secret), and a reliability dimension where every turn must follow a strict JSON schema and an illegal action is silently discarded. The engine is private and each match uses a fresh random map seed and opponent,...

    arxiv.org/abs/2606.24391 · PDF

  39. 39

    PHANTOM: A Large-Scale Dataset of Multimodal Adversarial Attacks for Vision-Language Models

    Simone Gallivanone, Hossein Khodadadi, Mauro Dore, Mauro Medda, Nicola Franco

    cs.AI · cs.LG

    We introduce a large-scale, open-source dataset of pre-generated adversarial attacks for vision-language models (VLMs). The dataset is designed to be diverse, representative, and practical, extending existing benchmarks by covering 10 high-level categories and 55 subcategories of harmful intents. Our primary goal is to make adversarial data accessible to the research community, given the computational cost and complexity of generating large...

    arxiv.org/abs/2606.24388 · PDF

  40. 40

    When Helpfulness Overrides Causal Caution: Context-Dependent Suppression and Recovery in LLMs

    Hiroshi Okumura

    cs.AI · cs.CY

    Large language models (LLMs) are increasingly integrated into decision-support roles in business and policy contexts. While prior benchmark studies have primarily evaluated LLMs' causal reasoning capabilities, a more fundamental epistemic dimension has been overlooked: Causal Caution, defined as the propensity to refrain from causal judgment when empirical evidence is insufficient. This study examines the systematic suppression of Causal...

    arxiv.org/abs/2606.24370 · PDF

  41. 41

    Accelerating Disaggregated RL for Visual Generative LLMs with Diffusion-Based Parallelism and Trainer-Assisted Generation

    Sijie Wang, Zhengyu Qing, Zhiqiang Tan, Yiming Yin, Yeqing Zhang, Yaoyuan Wang, Qiang Wang, Xiaowen Chu, Shaohuai Shi

    cs.AI · cs.DC · cs.NI · cs.PF

    Reinforcement learning (RL) has become a dominant post-training paradigm, driving the emergence of high-performance RL systems such as veRL for autoregressive large language models (LLMs). In parallel, diffusion-oriented RL algorithms, e.g., DanceGRPO and FlowGRPO, have rapidly expanded the scope of RL from language reasoning to diffusion-based visual and flow-based generation. However, efficient RL systems for diffusion generative LLMs...

    arxiv.org/abs/2606.24369 · PDF

  42. 42

    MVG-KAN: Multi-View Geo-Wind Guided KAN for PM$_{2.5}$ Forecasting

    Cheng Huang, Muyao Guan, Jairus Yougui Railey, Ning Xu, Honghui Xu, Changjiang Zhang, Zhen Zhang, Shiqing Zhang, Cong Bai

    cs.AI

    Accurate short-term PM$_{2.5}$ forecasting is important for public health protection, air-quality early warning, and urban environmental management. However, PM$_{2.5}$ variation is driven by multiple coupled factors, including stable periodic changes induced by human activities and meteorological regularity, station-specific short-term concentration evolution, and meteorology-driven pollutant dispersion among monitoring stations. Existing...

    arxiv.org/abs/2606.24347 · PDF

  43. 43

    Prob-BBDM: a Probabilistic Brownian Bridge Diffusion Model for MRI sequence image-to-image translation

    Martin Valls, Pascal Bourdon, Christine Fernandez-Maloigne, Guillaume Herpe, David Helbert

    cs.AI

    AI-driven image-to-image synthesis is rapidly advancing, with growing applications in medical imaging. Multi-modal image analysis plays a crucial role in optimizing examination quality, yet acquiring multiple imaging modalities in clinical settings remains resource-intensive and time-consuming, especially for 3D imaging. To address this challenge, we propose a novel image-to-image translation model based on Brownian Bridge Diffusion Models...

    arxiv.org/abs/2606.24313 · PDF

  44. 44

    LemonHarness Technical Report

    Kailong Ren, Fubo Sun, Jiachen Liu, Liu Yang, Zimo Yin, Jiaying Li, Congli Yin, Ming He, Yu Huo, Jiawei Liu, Zeping...

    cs.AI

    As large language model (LLM) agents are applied to longer tasks, they increasingly modify workspace state across multiple rounds of iteration. However, agents typically observe only tool outputs and log fragments, while the actual state changes occur in the file system. Without explicit workspace boundaries, state-changing operations such as file writes and temporary artifact generation may scatter changes across paths. Over time, these...

    arxiv.org/abs/2606.24311 · PDF

  45. 45

    Tractable Reasoning and Conjunctive Query Answering for Defeasible DL-Lite under Rational Closure

    Giovanni Casini, Umberto Straccia

    cs.AI

    In Description Logics (DLs), reasoning under Rational Closure (RC) is a well-known and widely accepted non-monotonic formalism to handle defeasible knowledge. In this paper, we study the application of RC to the core and horn variants of the DL-Lite family of lightweight description logics. We analyze both entitlement (instance checking) and Conjunctive Query (CQ) answering under RC. Our main contribution is providing a plug-in architecture...

    arxiv.org/abs/2606.24279 · PDF

  46. 46

    Probing the Misaligned Thinking Process of Language Models

    Kaiwen Zhou, Constantin Venhoff, Jonathan Michala, Xin Eric Wang, William Saunders

    cs.AI

    Large language models exhibit a growing range of misaligned behaviors such as strategic deception, sandbagging, and self-preservation. As they are increasingly deployed in high-stakes settings, it is critical to reliably detect such behaviors to ensure safe and responsible use. In this work, we propose to monitor misalignment by decomposing it into fine-grained cognitive processes -- misalignment indicators -- and detecting their presence in...

    arxiv.org/abs/2606.24251 · PDF

  47. 47

    Towards Federated Long-Tailed Graph Learning: An Energy-Guided Dual Decoupling Approach

    Lianshuai Guo, Zhongzheng Yuan, Xunkai Li, Meixia Qu, Wenyu Wang

    cs.AI

    Federated Graph Learning facilitates collaborative graph modeling across distributed clients while preserving data privacy. However, real-world data categories frequently exhibit long-tailed distributions. Such statistical scarcity severely degrades performance in two ways: it biases the global model toward majority classes, and it structurally isolates minority nodes by submerging them in heterophilic, head-dominated neighborhoods. While...

    arxiv.org/abs/2606.24237 · PDF

  48. 48

    SP-Mind: An Autonomous Reasoning Agent for Spatial Proteomics Analysis

    Yucheng Yuan, Yuanfeng Ji, Zhongxiao Li, Ruijiang Li

    cs.AI

    Spatial proteomics enables single-cell-resolution characterization of protein expression within tissue architecture, playing a critical role in understanding tumor microenvironments and guiding precision medicine. However, current analysis workflows remain fragmented, requiring expert manual orchestration of heterogeneous tools and limiting research scalability and reproducibility. We present SP-Mind, the first autonomous AI agent designed to...

    arxiv.org/abs/2606.24235 · PDF

  49. 49

    FlowR2A: Learning Reward-to-Action Distribution for Multimodal Driving Planning

    Xirui Li, Zhe Liu, Xiaoqing Ye, Wenhua Han, Yifeng Pan, Junyu Han, Hengshuang Zhao

    cs.AI

    Multimodal driving planning faces a long-standing tension between two paradigms: scoring-based methods benefit from dense reward supervision but are confined to a fixed action vocabulary, while anchor-based methods generate proposals dynamically yet suffer from sparse supervision constrained to a single ground-truth trajectory. In this work, we propose FlowR2A, which resolves this tension by reframing simulation-based rewards from...

    arxiv.org/abs/2606.24231 · PDF

  50. 50

    Exploring the relationship between human-centric AI and firm idiosyncratic risks

    Zhen-Yuan Ralph Liu, Yu-Ting Wang, Jia-Jia Yan, Shivam Gupta, Mihalis Giannakis

    cs.AI

    Despite the extensive discussions of human-centric AI (HCAI) in Industry 5.0, its effects on firms' idiosyncratic risks (IR) remains underexplored. This is an imperative issue for firms navigate financial risks during the current technological revolution, as IR reflects investor reactions to corporate heterogeneous AI strategies and implementations by isolating firm-level stock volatility from systematic factors. Integrating situated AI...

    arxiv.org/abs/2606.24224 · PDF

  51. 51

    Navigating User Behavior toward Personalized Multimodal Generation

    Hengji Zhou, Yufeng Liu, Ye Liu, Yong Xu, Lianghao Xia, Liqiang Nie

    cs.AI

    Modern AIGC pipelines deliver high-fidelity images and videos but presuppose a well-formed creation instruction, while end users rarely articulate visual details, leaving generators misaligned with user demand. We study personalized content generation, which turns a user's interaction history into an executable instruction for downstream synthesis, and identify two obstacles: behavior must be encoded in a form legible to language reasoning,...

    arxiv.org/abs/2606.24196 · PDF

  52. 52

    Data Scale, Not Latency, Shapes Cross-Lingual Encoder Transfer in Streaming ASR

    Nenad Banfic

    cs.AI

    Adapting a streaming speech recognition model to a new language requires choosing between two plausible warm starts: a multilingual (ML) encoder or an English-only (EN) encoder. The common intuition is that the multilingual encoder should help most at low data, but it is unclear how long that advantage persists, whether tight streaming latency amplifies it, and whether it survives deployment quantization. We answer these questions with a...

    arxiv.org/abs/2606.24169 · PDF

  53. 53

    An Introduction to Causal Reinforcement Learning

    Elias Bareinboim, Junzhe Zhang, Sanghack Lee

    cs.AI

    Causal inference provides a set of principles and tools that allow one to combine data and knowledge about an environment to reason with questions of counterfactual nature, i.e., what would have happened had reality been different, even when no data of this unrealized reality is currently available. Reinforcement learning provides methods to learn a policy that optimizes a specific measure (e.g., reward, regret) when the agent is deployed in...

    arxiv.org/abs/2606.24160 · PDF

  54. 54

    The Geometry Behind Diffusion and Flow Matching: Gradient Flows and Geodesics in Wasserstein Space

    Yian Yao, Weiwei Zhang

    cs.AI

    The space $\mathcal{P}_2(\mathbb{R}^d$) of probability measures with finite second moment carries a natural geometry: the quadratic Wasserstein distance W_2 makes it a complete metric space and, following Otto, a (formal) Riemannian manifold whose geodesics are the optimal-transport interpolations. On this manifold, the gradient flow of the free energy F(rho) = KL(rho || π) is exactly the Fokker-Planck equation, and its implicit-Euler...

    arxiv.org/abs/2606.24157 · PDF

  55. 55

    T2D-Bench: Evidence-Gated Evaluation of LLM Outputs for Type 2 Diabetes Using a Multi-Layer Clinical-Lifestyle Knowledge Graph

    Saba A. Farahani, Hung Cao, Ramesh Jain, Amir M. Rahmani

    cs.AI

    Large language models (LLMs) can produce clinically fluent recommendations for type 2 diabetes while failing to satisfy guideline constraints or explicitly justify lifestyle-related glycemic claims. We present T2D-Bench, a reproducible benchmark and evidence-gated evaluation framework for testing whether LLM outputs satisfy explicit, graph-checkable evidence requirements. T2D-Bench is built on a multi-layer clinical-lifestyle knowledge graph...

    arxiv.org/abs/2606.24145 · PDF

  56. 56

    OmniPath: A Multi-Modal Agentic Framework for Auditing Wheelchair Accessibility

    ASM Mobarak Hossain, Nadim Mahmud, Vaskar Raychoudhury, Md Osman Gani

    cs.AI

    For a wheelchair user, a standard blue line on a map is often a broken promise. While platforms like OpenStreetMap (OSM) successfully capture where a path is, they frequently fail to convey how it physically feels to travel on it. This information barrier is problematic for wheelchair users. To solve this issue, we present OmniPath, a system that moves from passive mapping to proactive environmental auditing. Our framework fuses the network...

    arxiv.org/abs/2606.24129 · PDF

  57. 57

    VeryTrace: Verifying Reasoning Traces through Compilable Formalism and Structured Verification

    Ninghan Zhong, Ahmet Ege Tanriverdi, Kaan Kale, Sriram Vishwanath

    cs.AI

    Multi-step reasoning with Chain-of-Thought (CoT) prompting remains fragile: logical errors or hallucinations in early steps silently propagate, producing confident but incorrect conclusions. This paper presents VeryTrace, a zero-shot verification-and-repair framework that formalizes natural-language reasoning traces into a structured, compilable representation. VeryTrace introduces a Domain-Specific Language (DSL) that (i) makes step...

    arxiv.org/abs/2606.24124 · PDF

  58. 58

    ReMMD: Realistic Multilingual Multi-Image Agentic Verification for Multimodal Misinformation Detection

    Chenhao Dang, Dantong Zhu, Jun Yang, Conghui He, Weijia Li

    cs.AI

    Multimodal misinformation detection is increasingly important because viral posts now combine long multilingual narratives, several images, mixed provenance, and subtle text--image framing errors. Existing benchmarks and methods remain poorly matched to this setting: they usually isolate short captions, single images, binary labels, or one manipulation source, while agentic verification remains costly under realistic evidence search. We...

    arxiv.org/abs/2606.24112 · PDF

  59. 59

    Exploring Academic Influence of Algorithms by Co-occurrence Network Based on Full-text of Academic Papers

    Yuzhuo Wang, Chengzhi Zhang, Min Song, Seong Deok Kim, Youngsoo Ko, Juhee Lee

    cs.AI · cs.CL · cs.DL · cs.IR

    Algorithms have become central to scientific research in the era of artificial intelligence (AI). Although algorithm mentions in papers are often used to indicate popularity and influence, existing studies usually evaluate individual algorithms in isolation and pay limited attention to the collective influence formed through their interconnections. This study constructs large-scale algorithm co-occurrence networks in natural language...

    arxiv.org/abs/2606.24099 · PDF

  60. 60

    Beyond Trajectory Imitation: Strategy-Guided Policy Optimization for LLM Reasoning

    Tianyuan Shi, Canbin Huang, Bei Li, Xin Chen, Xiaojun Quan, Jingang Wang, Qifan Wang

    cs.AI

    Distilling reasoning capabilities from strong to weak language models typically involves imitating specific solution trajectories, effectively transferring what to answer rather than how to reason. This trajectory-level imitation encourages memorization of instance-specific steps rather than acquisition of transferable problem-solving skills, limiting generalization to novel problems. We propose Strategy-Guided Policy Optimization (SGPO),...

    arxiv.org/abs/2606.24064 · PDF

  61. 61

    Ensemble Feature Selection and Harris Hawks Optimization for Explainable Mental Health Risk Prediction in Female Sex Workers

    Ahnaf Atef Choudhury, Md. Parvej Hoque Palash, Shahriar Siddique Ayon, Ramkrishna Saha, Abdullah Al Mamun

    cs.AI · cs.LG

    One of the significant mental health issues affecting female sex workers (FSWs) is mental disorders, especially depression. Exposure to violence, stigma, and economic hardship further increases their psychological risk. Current machine learning (ML) models are typically ineffective at capturing the high-dimensional and complex risk patterns that exist in this marginalized group. This paper suggests a hybrid predictive model that merges an...

    arxiv.org/abs/2606.24047 · PDF

  62. 62

    Breaking the Filter Bubble: A Semantic Pareto-DQN Framework for Multi-Objective Recommendation

    Cláudio Lúcio Do Val Lopes, Lucca Machado da Silva, André de Oliveira Brandão

    cs.AI

    Recommender systems often induce filter bubbles and semantic homogenization by monolithically optimizing for immediate user engagement. Standard single-objective models, including traditional Deep Q-Networks, are ill-equipped to navigate the trade-offs between platform retention and critical societal values like information diversity and provider fairness. To address these limitations, we introduce a multi-objective reinforcement learning...

    arxiv.org/abs/2606.24042 · PDF

  63. 63

    Can Language Model Agents be Helpful Circuit Explainers in Mechanistic Interpretability?

    Ayan Antik Khan, Harsh Kohli, Yuekun Yao, Huan Sun, Ziyu Yao

    cs.AI

    Mechanistic interpretability has made substantial progress in automatically localizing circuits, but explaining what localized components do remains labor-intensive and difficult to standardize. In this work, we study whether language model (LM) agents can assist with this explanation problem once a circuit has already been identified. We introduce AgenticInterpBench, a benchmark for circuit explanation built from 84 semi-synthetic...

    arxiv.org/abs/2606.24026 · PDF

  64. 64

    Reinforcement Learning Towards Broadly and Persistently Beneficial Models

    Akshay V. Jagadeesh, Rahul K. Arora, Khaled Saab, Ali Malik, Mikhail Trofimov, Foivos Tsimpourlas, Johannes...

    cs.AI · cs.CL

    As AI systems are deployed across increasingly diverse and high-stakes settings, model alignment must generalize beyond the tasks and domains seen during training. This is especially important for reinforcement learning (RL), which can introduce unexpected misalignment through reward hacking, deception, or other unintended strategies. We study whether RL on beneficial behavior, instantiated in realistic domains, can produce broad and...

    arxiv.org/abs/2606.24014 · PDF

  65. 65

    Safe and Generalizable Hierarchical Multi-Agent RL via Constraint Manifold Control

    Zihao Guo, Jianing Zhao, Ling Li, Hao Liang, Giuseppe Loianno, Yali Du

    cs.AI

    Multi-agent systems are widely used in safety-critical applications that require coordinated behavior under strict safety constraints. Existing approaches face a fundamental trade-off: learning-based methods achieve strong empirical performance but lack theoretical safety guarantees, while control-theoretic methods enforce safety but often lead to overly conservative and inefficient behaviors. We propose a hierarchical multi-agent...

    arxiv.org/abs/2606.24010 · PDF

  66. 66

    Critique of Agent Model

    Eric Xing, Mingkai Deng, Jinyu Hou

    cs.AI · cs.LG · cs.MA · cs.RO

    What is an agent? What constitutes agency? With the rise of Large Language Model (LLM) systems marketed as ``coding agents'', ``AI co-scientists'', and other ``agentic" tools that promise to drive up productivity, and at the same time, ``existential" concerns such as AI escaping human control with destructive power under a speculative ``machine agency" against humans, it has become essential to clarify where automation ends and agency begins,...

    arxiv.org/abs/2606.23991 · 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.