大模型推理能力

Extracting Actionability from Machine Learning Models by Sub-optimal Deterministic Planning

Qiang Lyu, Yixin Chen, Zhaorong Li, Zhicheng Cui, Ling Chen, Xing Zhang, Haihua Shen

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI、cs.LG

Comments 16 pages, 4 figures

Simple Regret Optimization in Online Planning for Markov Decision Processes

Zohar Feldman, Carmel Domshlak

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI、cs.LG

Reason Wide, Not Deep: Amortizing the Reasoning Premium into Distilled Skills

广泛推理，而非深度推理：将推理溢价分摊到提炼技能中

Agamdeep Singh, Srishti Gautam, Priyanshu Gupta, Nikita Mehrotra, Tanmay Bakshi, Sumit Gulwani

机构 * Microsoft（微软公司） 纠错

专题命中 规划推理 ：reasoning(title,abstract)；分类 cs.AI

AI总结 该研究提出将推理模式的重复分摊为跨任务提炼的技能，在四个智能体基准上大幅降低代币量的同时恢复多数推理性能，部分场景下优于推理模式。

Comments COLM 2026 Efficient Reasoning Workshop

Before Reasoning Can Fail: Pre-Evidence Procedural Failures in Agentic RAG

在推理失效之前：智能体检索增强生成（RAG）中的证据前程序失效

Daeyoung Roh, Donghee Han

机构 * KAIST（韩国科学技术院） 纠错

专题命中 规划推理 ：reasoning(title,abstract)；分类 cs.AI

AI总结 该研究针对智能体RAG系统在推理前的程序失效，提出将错误答案分为两类失效，评估Read-Gate机制，发现强制阅读可提升准确率，且证据收集应作为独立轨迹控制问题评估。

Comments 22 pages, 7 figures. Code: https://github.com/Noverse0/before-reasoning-fails

SAGA: Scene-Aware, Goal-Evolving Agents for Long-Horizon Strategy Game Planning

SAGA: 场景感知、目标演化的长时域CivRealm策略规划智能体

Tianyu Jin, Shuo Chen, Yida Wang, Liuyu Xiang, Yingzhuo Liu, Yexin Li, Peipei Li, Zhaofeng He

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

AI总结 针对长时域策略游戏中场景盲视、上下文溢出和跨游戏学习浅层化问题，提出SAGA多智能体框架，通过地图语义场景图、工具增强规划器和双时域反馈循环，在FreeCiv中取得最高文明评分并降低输出令牌27%。

Comments 18 pages, 4 figures. Code https://github.com/Kazecloudk/SAGA-Scene-Aware-Goal-Evolving-Agents-for-Long-Horizon-Strategy-Game-Planning

Automating RT Planning at Scale: High Quality Data For AI Training

大规模自动化放疗计划制定：为AI训练提供高质量数据

Riqiang Gao, Mamadou Diallo, Han Liu, Anthony Magliari, Jonathan Sackett, Wilko Verbakel, Sandra Meyers, Rafe Mcbeth, Masoud Zarepisheh, Simon Arberet, Martin Kraus, Florin C. Ghesu, Ali Kamen

专题命中 规划推理 ：planning(title,abstract)；分类 cs.LG

AI总结 本文提出了一种可扩展的自动化放疗计划制定系统，生成高质量的治疗计划，克服了AI驱动放疗计划发展中的关键障碍，其计划质量与人工生成的计划相当。

Comments radiotherapy planning, data for AI training

Comparing and Integrating Constraint Programming and Temporal Planning for Quantum Circuit Compilation

比较和整合约束编程与时间规划用于量子电路编译

Kyle E. C. Booth, Minh Do, J. Christopher Beck, Eleanor Rieffel, Davide Venturelli, Jeremy Frank

机构 * Quantum Artificial Intelligence Laboratory, NASA Ames Research Center（量子人工智能实验室，美国国家航空航天局阿姆斯特朗研究中心） ； Planning and Scheduling Group, NASA Ames Research Center（规划与调度组，美国国家航空航天局阿姆斯特朗研究中心） ； USRA Research Institute for Advanced Computer Science（美国宇航局高级计算机科学研究所） ； Stinger Ghaffarian Technologies, Inc.（Stinger Ghaffarian技术公司） ； Department of Mechanical & Industrial Engineering, University of Toronto（多伦多大学机械与工业工程系） 纠错

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

AI总结 本文比较了约束编程与时间规划在量子电路编译中的应用，提出混合方法提升求解质量，证明混合方法在多数问题中优于单独使用时间规划。

Comments 9 pages, 2 figures, Proceedings of the 28th International Conference of Automated Planning and Scheduling 2018 (ICAPS-18)

FocalAD: Local Motion Planning for End-to-End Autonomous Driving

FocalAD：端到端自动驾驶中的局部运动规划

Bin Sun, Boao Zhang, Jiayi Lu, Xinjie Feng, Jiachen Shang, Rui Cao, Mengchao Zheng, Chuanye Wang, Shichun Yang, Yaoguang Cao, Ziying Song

机构 * School of Transportation Science and Engineering（交通科学与工程学院） ； State Key Lab of Intelligent Transportation System（智能交通运输系统国家重点实验室） ； Hangzhou International Innovation Institute（杭州国际创新研究院） ； School of Computer Science and Technolog（计算机科学与技术学院） ； Beijing Jiaotong University（北京交通大学） 纠错

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

AI总结 本文提出FocalAD框架，通过增强局部运动表示和关注关键局部邻居，提升自动驾驶的规划可靠性与鲁棒性，在nuScenes和Bench2Drive等数据集上表现优异。

Journal ref FocalAD: Local Motion Planning for End-to-End Autonomous Driving. Automot. Innov. (2026)

LocationReasoner: Evaluating LLMs on Real-World Site Selection Reasoning

LocationReasoner: 评估大语言模型在真实世界选址推理中的能力

Miho Koda, Yu Zheng, Ruixian Ma, Mingyang Sun, Devesh Pansare, Fabio Duarte, Paolo Santi

专题命中 规划推理 ：reasoning(title,abstract)；分类 cs.AI

AI总结 本文提出LocationReasoner基准测试，评估大语言模型在复杂现实选址场景中的推理能力，发现先进模型在实际应用中表现有限，且代理策略常因过度推理而效果不佳。

Comments ICLR 2026 Workshop on Efficient Spatial Reasoning

Interpretable Responsibility Sharing as a Heuristic for Task and Motion Planning

可解释的责任共享作为任务与运动规划的启发式方法

Arda Sarp Yenicesu, Sepehr Nourmohammadi, Berk Cicek, Ozgur S. Oguz

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

AI总结 本文提出了一种名为可解释责任共享（IRS）的新型启发式方法，用于提升家用机器人任务与运动规划的效率，通过利用人类构建的环境和固有偏见，结合辅助物体简化任务执行。

Comments Accepted for the Special Issue "Planning and Learning for Autonomous Robotics" in Robotics and Autonomous Systems

MedEyes: Learning Dynamic Visual Focus for Medical Progressive Diagnosis

MedEyes: 学习动态视觉聚焦以进行医学逐步诊断

Chunzheng Zhu, Yangfang Lin, Shen Chen, Yijun Wang, Jianxin Lin

专题命中 规划推理 ：reasoning(abstract,comments)；chain-of-thought(abstract,comments)；CoT(abstract,comments)；分类 cs.AI

AI总结 MedEyes通过动态视觉聚焦和双模式探索策略，提升医学逐步诊断的准确性与临床相关性。

Comments AAAI 2026, Medical Chain-of-Thought (CoT), Reinforcement Learning with Verifiable Rewards (RLVR), Multimodal Grounded Reasoning

Automated Planning for Optimal Data Pipeline Instantiation

最优数据管道实例化的自动化规划

Leonardo Rosa Amado, Adriano Vogel, Dalvan Griebler, Gabriel Paludo Licks, Eric Simon, Felipe Meneguzzi

机构 * Pontifical Catholic University of Rio Grande do Sul, Brazil（里约格朗德杜斯鲁斯天主教大学） ； Johannes Kepler University Linz, Austria（林茨约翰·凯撒大学） ； Sapienza University of Rome, Italy（罗马萨皮恩扎大学） ； SAP Labs, France（SAP实验室） ； University of Aberdeen, Scotland（阿伯丁大学） 纠错

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

AI总结 本文提出了一种基于动作成本的规划方法，用于优化数据管道实例化，通过启发式算法减少总执行时间，并在实验中验证了其有效性。

Journal ref Proceedings of the ECAI Workshop on AI-based Planning for Complex Real-World Applications (CAIPI 2025)

Altruistic Maneuver Planning for Cooperative Autonomous Vehicles Using Multi-agent Advantage Actor-Critic

为合作自主车辆的利他性动作规划使用多智能体优势Actor-Critic

Behrad Toghi, Rodolfo Valiente, Dorsa Sadigh, Ramtin Pedarsani, Yaser P. Fallah

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

AI总结 本文提出一种多智能体优势Actor-Critic算法，用于自动驾驶车辆在混合交通环境中的利他性动作规划，以提升交通效率与安全。

Comments Accepted to 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2021) - Workshop on Autonomous Driving: Perception, Prediction and Planning

Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models

Vlad Sobal, Wancong Zhang, Kyunghyun Cho, Randall Balestriero, Tim G. J. Rudner, Yann LeCun

机构 * New York University（纽约大学） ； Genentech（基因泰克） ； Brown University（布朗大学） ； University of Toronto（多伦多大学） ； Meta – FAIR 纠错

专题命中 规划推理 ：planning(title,abstract)；分类 cs.LG

Comments Project web page: https://latent-planning.github.io/

NL2Plan: Robust LLM-Driven Planning from Minimal Text Descriptions

Elliot Gestrin, Marco Kuhlmann, Jendrik Seipp

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments Accepted for the ICAPS 2024 Workshop on Human-Aware and Explainable Planning

Urban Comfort Assessment in the Era of Digital Planning: A Multidimensional, Data-driven, and AI-assisted Framework

Sijie Yang, Binyu Lei, Filip Biljecki

机构 * Department of Architecture, National University of Singapore（新加坡国立大学建筑系） ； School of Engineering and Applied Science, University of Pennsylvania（宾夕法尼亚大学工程与应用科学学院） ； Department of Real Estate, National University of Singapore（新加坡国立大学房地产系） 纠错

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments Presented at 19th International Conference on Computational Urban Planning and Urban Management (CUPUM 2025)

Planning with Adaptive World Models for Autonomous Driving

Arun Balajee Vasudevan, Neehar Peri, Jeff Schneider, Deva Ramanan

专题命中 规划推理 ：planning(title,abstract)；分类 cs.LG

Comments This project has been accepted to the International Conference on Robotics and Automation (ICRA) 2025. Project Page: https://arunbalajeev.github.io/world_models_planning/world_model_paper.html

AAAI Workshop on AI Planning for Cyber-Physical Systems -- CAIPI24

Oliver Niggemann, Gautam Biswas, Alexander Diedrich, Jonas Ehrhardt, René Heesch, Niklas Widulle

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments This is the Proceedings of the AAAI Workshop on AI Planning for Cyber-Physical Systems - CAIPI24, which was held in Vancouver, CA, February 26, 2024

On the Prospects of Incorporating Large Language Models (LLMs) in Automated Planning and Scheduling (APS)

Vishal Pallagani, Kaushik Roy, Bharath Muppasani, Francesco Fabiano, Andrea Loreggia, Keerthiram Murugesan, Biplav Srivastava, Francesca Rossi, Lior Horesh, Amit Sheth

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Journal ref Proceedings of the International Conference on Automated Planning and Scheduling, 34(1), 432-444 (2024)

Epistemic Exploration for Generalizable Planning and Learning in Non-Stationary Settings

Rushang Karia, Pulkit Verma, Alberto Speranzon, Siddharth Srivastava

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments To appear at ICAPS-24

Journal ref Proceedings of the International Conference on Automated Planning and Scheduling, 34(1), 310-318, 2024

A Planning Ontology to Represent and Exploit Planning Knowledge for Performance Efficiency

Bharath Muppasani, Vishal Pallagani, Biplav Srivastava, Raghava Mutharaju, Michael N. Huhns, Vignesh Narayanan

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments Ontology, Automated Planning, Planner Improvement

Human-Centric Goal Reasoning with Ripple-Down Rules

Kenji Brameld, Germán Castro, Claude Sammut, Mark Roberts, David W. Aha

专题命中 规划推理 ：reasoning(title,abstract)；分类 cs.AI

Comments Proceedings of the Ninth Goal Reasoning Workshop (Advances in Cognitive Systems, 2021)

Planning Landmark Based Goal Recognition Revisited: Does Using Initial State Landmarks Make Sense?

Nils Wilken, Lea Cohausz, Christian Bartelt, Heiner Stuckenschmidt

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments Full publication: Wilken, N., Cohausz, L., Bartelt, C., Stuckenschmidt, H. (2023). Planning Landmark Based Goal Recognition Revisited: Does Using Initial State Landmarks Make Sense?. In: Seipel, D., Steen, A. (eds) KI 2023: Advances in Artificial Intelligence. KI 2023. Lecture Notes in Computer Science(), vol 14236. Springer, Cham. arXiv admin note: text overlap with arXiv:2301.10571

Planning for Manipulation among Movable Objects: Deciding Which Objects Go Where, in What Order, and How

Dhruv Saxena, Maxim Likhachev

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments Accepted for publication at the International Conference on Automated Planning and Scheduling (ICAPS), 2023

Deliberative Acting, Online Planning and Learning with Hierarchical Operational Models

Sunandita Patra, James Mason, Malik Ghallab, Dana Nau, Paolo Traverso

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments Published in Artificial Intelligence (AIJ). Please cite as: Sunandita Patra, James Mason, Malik Ghallab, Dana Nau, Paolo Traverso. Deliberative Acting, Planning and Learning with Hierarchical Operational Models. Artificial Intelligence, Elsevier, 2021, 299, pp.103523. 10.1016/j.artint.2021.103523. arXiv admin note: text overlap with arXiv:2003.03932

Journal ref Artificial Intelligence, Elsevier, 2021, 299, pp.103523

Approximate Computing for Robotic path planning -- Experimentation, Case Study and Practical Implications

Hrishav Bakul Barua

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments Approximate Computing, Multi-robot Systems, Multi-agent Systems, Good Enough Computing, Green Computing, Robot Path Planning, Energy-Efficient Computing

Encoding Compositionality in Classical Planning Solutions

Angeline Aguinaldo, William Regli

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments IJCAI Generalization in Planning Workshop 2021

Guiding Robot Exploration in Reinforcement Learning via Automated Planning

Yohei Hayamizu, Saeid Amiri, Kishan Chandan, Keiki Takadama, Shiqi Zhang

专题命中 规划推理 ：planning(title,abstract)；分类 cs.LG

Comments Accepted in International Conference of Planning and Scheduling (ICAPS-21)

Iterative Planning with Plan-Space Explanations: A Tool and User Study

Rebecca Eifler, Jörg Hoffmann

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI

Comments Proceedings of the International Workshop of Explainable AI Planning (XAIP'20), at ICAPS'20

On the Importance of Domain Model Configuration for Automated Planning Engines

Mauro Vallati, Lukas Chrpa, Thomas L. McCluskey, Frank Hutter

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI；reasoning(comments)

Comments Under consideration in Journal of Automated Reasoning