JAIR

Average Reward Reinforcement Learning for Omega-Regular and Mean-Payoff Objectives

用于Ω-正则和平均收益目标的平均收益强化学习

Milad Kazemi, Mateo Perez, Fabio Somenzi, Sadegh Soudjani, Ashutosh Trivedi, Alvaro Velasquez

机构 * King’s College London（伦敦大学国王学院） ； University of Colorado Boulder（科罗拉多大学波尔德分校） ； University of Birmingham, UK（英国伯明翰大学） ； MPI-SWS（马克斯·普朗克研究所（SWS）） 纠错

AI总结 本文提出首个无模型强化学习框架，将绝对活性规范转化为平均收益目标，支持未知通信马尔可夫决策过程的学习，且在无episodic重置情况下收敛。

Comments 29 pages

Journal ref Journal of Artificial Intelligence Research 85, Article 25(March 2026)

$\aleph$-IPOMDP: Mitigating Deception in a Cognitive Hierarchy with Off-Policy Counterfactual Anomaly Detection

$\aleph$-IPOMDP：通过非策略反事实异常检测缓解认知层级中的欺骗

Nitay Alon, Joseph M. Barnby, Stefan Sarkadi, Lion Schulz, Jeffrey S. Rosenschein, Peter Dayan

AI总结 $\aleph$-IPOMDP通过非策略反事实异常检测机制，缓解认知层级中的欺骗问题，提升博弈中的公平性和安全性。

Comments 28 pages, 12 figures

Journal ref Journal of Artificial Intelligence Research (JAIR), Volume 85, Article 14, 2026

The Winnability of Klondike Solitaire and Many Other Patience Games

克朗迪克纸牌游戏及其他许多耐心游戏的可赢性

Charlie Blake, Ian P. Gent

机构 * University of St Andrews（圣安德鲁大学） 纠错

AI总结 Solvitaire通过深度优先搜索等技术，确定了73种单人纸牌游戏的可赢百分比，显著提升了对克朗迪克等游戏的可赢性理解。

Comments Published version, please cite the JAIR version published in Feb 2026, doi 10.1613/jair.1.17167

Journal ref Journal of ArtificialIntelligence Research, Volume 85, Article 21 (February 2026), 47 pages

Procedural Fairness in Machine Learning

机器学习中的程序公平性

Ziming Wang, Changwu Huang, Ke Tang, Xin Yao

机构 * Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation, Department of Computer Science and Engineering, Southern University of Science and Technology（脑启发智能计算广东省重点实验室，计算机科学与工程系，南方科技大学） ； School of AI and Liberal Arts, Beijing Normal-Hong Kong Baptist University（人工智能与人文学院，北京师范大学-香港 Baptist大学） ； School of Data Science, Lingnan University（数据科学学院，岭南大学） 纠错

AI总结 本文提出了一种评估机器学习模型程序公平性的新度量标准GPF_FAE，并通过实验展示了如何识别导致程序不公平性的特征，从而提升模型的程序和分配公平性。

Comments 30 pages, 14 figures, Published in JAIR

Journal ref Journal of Artificial Intelligence Research 85, Article 20 (February 2026)

Homotopy-Aware Multi-Agent Path Planning on Plane

平面中具有同伦意识的多智能体路径规划

Kazumi Kasaura

AI总结 本文提出基于Dynnikov坐标的高效框架，用于平面中多智能体路径规划，通过结合优先规划生成多个同伦不同的解决方案，并验证其在避免局部最优解方面的有效性。

Comments 23 pages with 5 pages of references and appendices, 25 figures

Journal ref Journal of Artificial Intelligence Research 85, Article 15 (February 2026)

T-COL: Generating Counterfactual Explanations for General User Preferences on Variable Machine Learning Systems

T-COL：为可变机器学习系统生成通用用户偏好的反事实解释

Ming Wang, Daling Wang, Wenfang Wu, Shi Feng, Yifei Zhang

机构 * Northeastern University（东北大学） 纠错

AI总结 T-COL 通过基于树的条件可选链接方法，为可变机器学习系统生成适应一般用户偏好的反事实解释，提升解释的鲁棒性和适应性。

Journal ref Journal of Artificial Intelligence Research, Vol. 85 (2026)

Asymptotically Fair and Truthful Allocation of Public Goods

公共物品的渐近公平和 truthful 分配

Pouya Kananian, Arnesh Sujanani, Seyed Majid Zahedi

AI总结 本文提出PPGA算法，通过引入差分隐私实现公共物品分配的渐近公平和 truthful，解决了传统方法在大规模场景下的局限性。

Comments 26 pages, typos corrected, section 4.3 added, new author added

Journal ref JAIR 85 (2026) 1-27

Autotelic Agents with Intrinsically Motivated Goal-Conditioned Reinforcement Learning: a Short Survey

自主机器的自激励目标导向强化学习：一种简短综述

Cédric Colas, Tristan Karch, Olivier Sigaud, Pierre-Yves Oudeyer

AI总结 本文提出了一种基于目标导向强化学习的发展性强化学习框架，旨在解决自激励技能获取问题，并探讨了目标表示和优先级学习的挑战。

Journal ref JAIR 2022

Probabilistically Tightened Linear Relaxation-based Perturbation Analysis for Neural Network Verification

基于概率紧缩的线性松弛扰动分析用于神经网络验证

Luca Marzari, Ferdinando Cicalese, Alessandro Farinelli

机构 * Department of Computer Science, University of Verona（计算机科学系，威尼斯大学） 纠错

AI总结 PT-LiRPA通过概率紧缩线性松弛扰动分析提升神经网络验证的鲁棒性证书，显著提高验证正确性的概率保证。

Comments Accepted at the Journal of Artificial Intelligence Research (JAIR)

Journal ref Journal of Artificial Intelligence Research 84, Article 30 (December 2025), 34 pages

Agentic Large Language Models, a survey

代理大语言模型：综述

Aske Plaat, Max van Duijn, Niki van Stein, Mike Preuss, Peter van der Putten, Kees Joost Batenburg

机构 * Leiden University Leiden Netherlands ； Leiden University \& AI Lab, Pegasystems Leiden Netherlands ； Leiden University ； Leiden University \& AI Lab, Pegasystems 纠错

AI总结 本文综述了代理大语言模型的研究现状，探讨了其在医疗诊断、物流和金融分析等领域的应用，并提出通过推理、行动和交互提升大语言模型能力的未来研究方向。

Comments Website: https://askeplaat.github.io/agentic-llm-survey-site/

Journal ref JAIR volume 84, article 29, December 2025

Mechanisms of Symbol Processing for In-Context Learning in Transformer Networks

Transformer网络中上下文学习符号处理的机制

Paul Smolensky, Roland Fernandez, Zhenghao Herbert Zhou, Mattia Opper, Adam Davies, Jianfeng Gao

AI总结 本文提出了一种基于生产系统语言PSL的转换器架构，用于提升转换器在符号处理中的能力，展示了其在抽象符号任务中的应用和图灵通用性。

Journal ref Journal of Artificial Intelligence Research, 84(23) 2025

Generating Streamlining Constraints with Large Language Models

Florentina Voboril, Vaidyanathan Peruvemba Ramaswamy, Stefan Szeider

机构 * Algorithms and Complexity Group TU Wien（算法与复杂性组维也纳技术大学） 纠错

Comments 23 page; deeper analysis of streamliners and statistics about benchmark instances added

Journal ref F. Voboril, V. P. Ramaswamy, S. Szeider, Generating Streamlining Constraints with Large Language Models, Journal of Artificial Intelligence Research, volume 84, pages 16:1-16:19, 2025

Trustworthy Transfer Learning: A Survey

Jun Wu, Jingrui He

机构 * Michigan State University（密歇根州立大学） ； University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校） 纠错

Comments Accepted by Journal of Artificial Intelligence Research (JAIR)

Journal ref Journal of Artificial Intelligence Research 84, Article 20(November 2025), 59 pages

Regret Analysis for Randomized Gaussian Process Upper Confidence Bound

Shion Takeno, Yu Inatsu, Masayuki Karasuyama

机构 * Nagoya University（名古屋大学） ； RIKEN AIP（理化学研究所AIP） ； Nagoya Institute of Technology（名古屋工业大学） 纠错

Comments 37 pages, 4 figures. Accepted to Journal of Artificial Intelligence Research as an extended paper from arXiv:2302.01511

Journal ref Journal of Artificial Intelligence Research 84, Article 18, (November 2025)

Guided by Stars: Interpretable Concept Learning Over Time Series via Temporal Logic Semantics

Irene Ferfoglia, Simone Silvetti, Gaia Saveri, Laura Nenzi, Luca Bortolussi

机构 * Università degli Studi di Trieste（特里埃斯特大学） 纠错

Comments submitted to Journal of Artificial Intelligence Research (JAIR), 2025

Representative Social Choice: From Learning Theory to AI Alignment

Tianyi Qiu

机构 * Peking University（北京大学） ； UC Berkeley（加州大学伯克利分校） ； Center for Human-Compatible AI（人类兼容人工智能中心） 纠错

Comments Journal of Artificial Intelligence Research, in press. Best Paper at NeurIPS 2024 Pluralistic Alignment Workshop

Thousands of AI Authors on the Future of AI

Katja Grace, Harlan Stewart, Julia Fabienne Sandkühler, Stephen Thomas, Ben Weinstein-Raun, Jan Brauner, Richard C. Korzekwa

机构 * AI Impacts ； Berkeley, California United States（伯克利加州大学） ； Department of Psychology University of Bonn（波恩大学心理学系） ； University of Bonn（波恩大学） ； Department of Computer Science University of Oxford（牛津大学计算机科学系） ； University of Oxford（牛津大学） ； Independent（独立研究者） 纠错

Comments The asterisk indicates the corresponding author. The dagger indicates equal contribution

Journal ref Journal of Artificial Intelligence Research 84:9 (2025)

A Survey on Data Selection for LLM Instruction Tuning

Bolin Zhang, Jiahao Wang, Qianlong Du, Jiajun Zhang, Zhiying Tu, Dianhui Chu

机构 * Harbin Institute of Technology（哈尔滨工业大学） ； Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所） ； Harbin Institute of Technology (Weihai)（哈尔滨工业大学（威海）） 纠错

Comments Published in JAIR (Vol. 83, Article 32, 2025)

Journal ref Journal of Artificial Intelligence Research, 83:32, 2025

An MRP Formulation for Supervised Learning: Generalized Temporal Difference Learning Models

Yangchen Pan, Junfeng Wen, Chenjun Xiao, Philip Torr

机构 * University of Oxford（牛津大学） ； Carleton University（卡尔顿大学） ； The Chinese University of Hong Kong (Shenzhen)（香港中文大学（深圳）） 纠错

Comments Accepted by JAIR. The abstract above is more concise than the one in the paper to meet the requirements of the arXiv website

Implicit Safe Set Algorithm for Provably Safe Reinforcement Learning

Weiye Zhao, Feihan Li, Changliu Liu

机构 * Carnegie Mellon University（卡内基梅隆大学） 纠错

Comments Accepted to Journal of Artificial Intelligence Research. arXiv admin note: text overlap with arXiv:2308.13140

Learning How to Vote with Principles: Axiomatic Insights Into the Collective Decisions of Neural Networks

Levin Hornischer, Zoi Terzopoulou

机构 * Munich Center for Mathematical Philosophy, LMU Munich Munich Germany ； GATE, CNRS, Universit\'e Jean Monnet, Universit\'e Lumiere Lyon 2 Saint-Etienne France ； Munich Center for Mathematical Philosophy, LMU Munich ； GATE, CNRS, Universit\'e Jean Monnet, Universit\'e Lumiere Lyon 2 纠错

Comments 44 pages, 21 figures, 14 tables. Updated and published version

Journal ref Journal of Artificial Intelligence Research 83, Article 25 (August 2025)

Scalable Synthesis of Formally Verified Neural Value Function for Hamilton-Jacobi Reachability Analysis

Yujie Yang, Hanjiang Hu, Tianhao Wei, Shengbo Eben Li, Changliu Liu

Journal ref Journal of Artificial Intelligence Research 83 (2025) 19

The Search for Stability: Learning Dynamics of Strategic Publishers with Initial Documents

Omer Madmon, Idan Pipano, Itamar Reinman, Moshe Tennenholtz

Comments Published in the Journal of Artificial Intelligence Research 83 (2025)

Journal ref Journal of Artificial Intelligence Research 83 (2025)

Towards Generalist Robot Learning from Internet Video: A Survey

Robert McCarthy, Daniel C. H. Tan, Dominik Schmidt, Fernando Acero, Nathan Herr, Yilun Du, Thomas G. Thuruthel, Zhibin Li

机构 * University College London（伦敦大学学院） ； Massachusetts Institute of Technology（麻省理工学院） 纠错

Journal ref Journal of Artificial Intelligence Research, Vol.83, Article 12. Publication date: July 2025

On CNF Conversion for SAT and SMT Enumeration

Gabriele Masina, Giuseppe Spallitta, Roberto Sebastiani

Comments to be published in the Journal of Artificial Intelligence Research. 26 pages, 4 figures (48 pages, 19 figures with appendix)

Journal ref Journal of Artificial Intelligence Research 83 (2025)

DSAC: Distributional Soft Actor-Critic for Risk-Sensitive Reinforcement Learning

Xiaoteng Ma, Junyao Chen, Li Xia, Jun Yang, Qianchuan Zhao, Zhengyuan Zhou

机构 * Department of Automation, Tsinghua University（清华大学自动化系） ； School of Engineering and Applied Science, Columbia University（哥伦比亚大学工程与应用科学学院） ； School of Business, Sun Yat-sen University（中山大学商学院） ； Stern School of Business, New York University（纽约大学斯特恩商学院） 纠错

Comments Accecpted by Journal of Artificial Intelligence Research

TabID: Automatic Identification and Tabulation of Subproblems in Constraint Models

Özgür Akgün, Ian P. Gent, Christopher Jefferson, Zeynep Kiziltan, Ian Miguel, Peter Nightingale, András Z. Salamon, Felix Ulrich-Oltean

机构 * School of Computer Science, University of St Andrews（圣安德鲁大学计算机科学学院） ； School of Science and Engineering, University of Dundee（邓迪大学科学与工程学院） ； Department of Computer Science, University of York（约克大学计算机科学系） 纠错

Journal ref Journal of Artificial Intelligence Research 82 (2025) 1999-2056

WorldView-Bench: A Benchmark for Evaluating Global Cultural Perspectives in Large Language Models

Abdullah Mushtaq, Imran Taj, Rafay Naeem, Ibrahim Ghaznavi, Junaid Qadir

机构 * Information Technology University Department of Computer Science（信息科技大学计算机科学系） ； Zayed University College of Interdisciplinary Studies（扎耶德大学跨学科研究学院） ； Computer Science and Engineering Department（计算机科学与工程系） 纠错

Comments Preprint. Submitted to the Journal of Artificial Intelligence Research (JAIR) on April 29, 2025

AI Governance in the GCC States: A Comparative Analysis of National AI Strategies

Mohammad Rashed Albous, Odeh Rashed Al-Jayyousi, Melodena Stephens

Comments 33 pages,6 figures, 11 tables

Journal ref Journal of Artificial Intelligence Research, 82, 2389-2422 (2025)

CADS: A Systematic Literature Review on the Challenges of Abstractive Dialogue Summarization

Frederic Kirstein, Jan Philip Wahle, Bela Gipp, Terry Ruas

机构 * Georg-August University（哥廷根大学） 纠错

Comments Published in the Journal of Artificial Intelligence Research (JAIR) (https://www.jair.org/index.php/jair/article/view/16674)

Journal ref Journal of Artificial Intelligence Research (JAIR), Vol. 82, 2025