大模型推理能力

Evaluating the Safety and Skill Reasoning of Large Reasoning Models Under Compute Constraints

Adarsha Balaji, Le Chen, Rajeev Thakur, Franck Cappello, Sandeep Madireddy

机构 * Argonne National Laboratory（阿贡国家实验室） ； Mathematics and Computer Science Division（数学与计算机科学 division） ； Data Science and Learning Division（数据科学与学习 division） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；test-time compute(abstract)

Analogy-Driven Financial Chain-of-Thought (AD-FCoT): A Prompting Approach for Financial Sentiment Analysis

Anmol Singhal Navya Singhal

机构 * New York University（纽约大学） ； University of Texas at Austin（德克萨斯大学奥斯汀分校） 纠错

专题命中 逻辑推理 ：chain-of-thought(title,abstract)；reasoning(abstract)；CoT(abstract)；logical reasoning(abstract)

Comments IEEE AIxB 2025

A Study on Individual Spatiotemporal Activity Generation Method Using MCP-Enhanced Chain-of-Thought Large Language Models

Yu Zhang, Yang Hu, De Wang

专题命中 逻辑推理 ：chain-of-thought(title,abstract)；reasoning(abstract)；CoT(abstract)；planning(abstract)

Chain-of-Thought for Autonomous Driving: A Comprehensive Survey and Future Prospects

Yixin Cui, Haotian Lin, Shuo Yang, Yixiao Wang, Yanjun Huang, Hong Chen

机构 * School of Automotive Studies, Tongji University（同济大学汽车学院） ； School of Physics Science and Engineering, Tongji University（同济大学物理科学与工程学院） ； Frontiers Science Center for Intelligent Autonomous Systems（智能自主系统前沿科学中心） ； College of Electronics and Information Engineering（电子信息学院） ； Clean Energy Automotive Engineering Center（清洁能源汽车工程中心） 纠错

专题命中 逻辑推理 ：chain-of-thought(title,abstract)；reasoning(abstract)；CoT(abstract)；logical reasoning(abstract)

Comments 18 pages, 6 figures

Cantor: Inspiring Multimodal Chain-of-Thought of MLLM

Timin Gao, Peixian Chen, Mengdan Zhang, Chaoyou Fu, Yunhang Shen, Yan Zhang, Shengchuan Zhang, Xiawu Zheng, Xing Sun, Liujuan Cao, Rongrong Ji

专题命中 逻辑推理 ：chain-of-thought(title,abstract)；reasoning(abstract)；CoT(abstract)；logical reasoning(abstract)

Comments The project page is available at https://ggg0919.github.io/cantor/

AS-ES Learning: Towards Efficient CoT Learning in Small Models

Nuwa Xi, Yuhan Chen, Sendong Zhao, Haochun Wang, Bing Qin, Ting Liu

专题命中 逻辑推理 ：CoT(title,abstract)；reasoning(abstract)；chain-of-thought(abstract)；logical reasoning(abstract)

Assessing Step-by-Step Reasoning against Lexical Negation: A Case Study on Syllogism

Mengyu Ye, Tatsuki Kuribayashi, Jun Suzuki, Goro Kobayashi, Hiroaki Funayama

专题命中 逻辑推理 ：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；logical reasoning(abstract)

Thinking Like an Expert:Multimodal Hypergraph-of-Thought (HoT) Reasoning to boost Foundation Modals

Fanglong Yao, Changyuan Tian, Jintao Liu, Zequn Zhang, Qing Liu, Li Jin, Shuchao Li, Xiaoyu Li, Xian Sun

专题命中 逻辑推理 ：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；logical reasoning(abstract)

Reasoning Depth and Environment Complexity: A Controlled Study of RLVR Data Allocation across Logical Reasoning Tasks

推理深度与环境复杂度：逻辑推理任务中RLVR数据分配的受控研究

Yihua Zhu, Qianying Liu, Fei Cheng, Jiaxin Wang, Akiko Aizawa, Sadao Kurohashi, Hidetoshi Shimodaira

机构 * Kyoto University（京都大学） ； University of Tokyo（东京大学） ； NII LLMC（日本信息处理学会LLMC） ； RIKEN（理化学研究所） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.CL、cs.AI

AI总结 通过将推理空间划分为深度和复杂度两个维度，并考虑四种推理形式，在合成知识图谱环境中进行受控实验，发现联合深度-复杂度覆盖优于单轴策略，不同推理家族对RLVR覆盖的反应非均匀，且均匀混合优于分阶段课程。

Comments Pre-print

Emergent Analogical Reasoning in Large Language Models

Taylor Webb, Keith J. Holyoak, Hongjing Lu

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.CL、cs.AI

Comments Published at Nature Human Behaviour (2023) https://doi.org/10.1038/s41562-023-01659-w

Symbolic-Neural Soft-Logic Reasoning: Towards Robust and Verifiable Thinking Chains via Cooperative Evolution

符号-神经软逻辑推理：通过协同进化实现鲁棒且可验证的思维链

Rui Wang, Zeming Wei, Yihao Zhang, Xiaokun Luan

专题命中 逻辑推理 ：reasoning(title,abstract)；CoT(abstract,abstract_cn)；chain-of-thought(abstract)

AI总结 提出符号-神经软逻辑推理（SSR）框架，通过结合LLM与符号推理并放松严格逻辑确定性，提升推理鲁棒性和可验证性，实验表明优于现有方法。

Learning When to Look: A Disentangled Curriculum for Strategic Perception in Multimodal Reasoning

学习何时观察：一种解耦的课程学习框架，用于多模态推理中的战略感知

Siqi Yang, Zilve Gao, Haibo Qiu, Fanfan Liu, Peng Shi, Zhixiong Zeng, Qingmin Liao, Lin Ma

机构 * Meituan（美团） ； Tsinghua University（清华大学） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；logical reasoning(abstract)

AI总结 本文提出了一种解耦课程学习框架，通过分阶段训练提升多模态推理中的抽象推理与战略视觉感知能力。

QuestBench: Can LLMs ask the right question to acquire information in reasoning tasks?

Belinda Z. Li, Been Kim, Zi Wang

机构 * Massachusetts Institute of Technology（麻省理工学院） ； Google DeepMind（谷歌DeepMind） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；planning(abstract)；logical reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Code and dataset are available at \url{https://github.com/google-deepmind/questbench}

Journal ref 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Track on Datasets and Benchmarks

FLARE: Faithful Logic-Aided Reasoning and Exploration

Erik Arakelyan, Pasquale Minervini, Pat Verga, Patrick Lewis, Isabelle Augenstein

机构 * University of Copenhagen（哥本哈根大学） ； University of Edinburgh（爱丁堡大学） ； Cohere ； NVIDIA 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Published at EMNLP 2025

RACE-Align: Retrieval-Augmented and Chain-of-Thought Enhanced Preference Alignment for Large Language Models

Qihang Yan, Xinyu Zhang, Luming Guo, Qi Zhang, Feifan Liu

机构 * ShanghaiTech University（上海科技大学） ； Henan University（河南大学） ； Liaoning University of Traditional Chinese Medicine（辽宁中医药大学） 纠错

专题命中 逻辑推理 ：chain-of-thought(title,abstract)；reasoning(abstract)；CoT(abstract)；分类 cs.CL、cs.AI、cs.LG

Automatic Curriculum Expert Iteration for Reliable LLM Reasoning

Zirui Zhao, Hanze Dong, Amrita Saha, Caiming Xiong, Doyen Sahoo

专题命中 逻辑推理 ：reasoning(title,abstract)；planning(abstract)；logical reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments 20 pages

Auto-Evolve: Enhancing Large Language Model's Performance via Self-Reasoning Framework

Krishna Aswani, Huilin Lu, Pranav Patankar, Priya Dhalwani, Iris Tan, Jayant Ganeshmohan, Simon Lacasse

专题命中 逻辑推理 ：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Accepted at EMNLP 2024

RIPPLECOT: Amplifying Ripple Effect of Knowledge Editing in Language Models via Chain-of-Thought In-Context Learning

Zihao Zhao, Yuchen Yang, Yijiang Li, Yinzhi Cao

专题命中 逻辑推理 ：chain-of-thought(title,abstract)；reasoning(abstract)；CoT(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments EMNLP findings

Advancing LLM Reasoning Generalists with Preference Trees

Lifan Yuan, Ganqu Cui, Hanbin Wang, Ning Ding, Xingyao Wang, Jia Deng, Boji Shan, Huimin Chen, Ruobing Xie, Yankai Lin, Zhenghao Liu, Bowen Zhou, Hao Peng, Zhiyuan Liu, Maosong Sun

专题命中 逻辑推理 ：reasoning(title,abstract)；planning(abstract)；logical reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Models and data are available at https://github.com/OpenBMB/Eurus

Structured Abductive-Deductive-Inductive Reasoning for LLMs via Algebraic Invariants

通过代数不变量实现LLMs的结构归纳-演绎-演绎推理

Sankalp Gilda, Shlok Gilda

机构 * DeepThought Solutions（深思科技） ； Department of Computer Science University of Florida（佛罗里达大学计算机科学系） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；chain-of-thought(abstract)；logical reasoning(abstract)；分类 cs.AI、cs.LG

AI总结 本文提出通过代数不变量实现LLMs的结构归纳-演绎-演绎推理框架，通过五种代数不变量确保推理链中结论的可靠性，防止逻辑不一致的累积。

Comments 10 pages + 3 pages references. Accepted as a poster at the ICLR 2026 Workshop for LLM Reasoning

Lost in the Logic: An Evaluation of Large Language Models' Reasoning Capabilities on LSAT Logic Games

Saumya Malik

专题命中 逻辑推理 ：reasoning(title,abstract)；chain-of-thought(abstract)；logical reasoning(abstract)；分类 cs.CL、cs.AI

Comments Bachelor's thesis. Dataset available on huggingface: https://huggingface.co/datasets/saumyamalik/lsat_logic_games-analytical_reasoning

Combining Large Language Models and Symbolic Reasoning for Multi-Robot Temporal Planning through Explainable Knowledge Bases

结合大语言模型与符号推理，通过可解释知识库实现多机器人时序规划

Enrico Saccon, Matteo Saveriano, Edoardo Lamon, Luigi Palopoli, Marco Roveri

专题命中 逻辑推理 ：planning(title,abstract)；reasoning(title)；分类 cs.AI

AI总结 本研究提出PLANTOR框架，结合LLM与符号推理，通过可解释知识库实现多机器人时序规划，在基准场景及多臂装配场景验证，可减少手动建模但需人工修正，主张混合工作流。

Biological Reasoning-Informed Regression for Interpretable Regulatory DNA Activity Prediction

面向可解释调控DNA活性预测的生物学推理引导回归

Yi Duan, Zhao Yang, Jiwei Zhu, Ying Ba, Chuan Cao, Bing Su

机构 * Gaoling School of Artificial Intelligence（甘岭人工智能学院） ； Renmin University of China（中国人民大学） ； Zhongguancun Academy（中关村学院） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.LG

AI总结 提出R3LM框架，通过结构化生物学知识教LLM进行推理引导回归，在增强子预测上达到最优性能并提供可解释机制。

Comments Accepted at KDD 2026 AI4Sciences Track

BioBridge: Bridging Proteins and Language for Enhanced Biological Reasoning with LLMs

BioBridge: 蛋白质与语言的桥梁：利用大语言模型增强生物推理

Yujia Wang, Jihong Guan, Wengen Li, Shuigeng Zhou, Xuhong Wang

机构 * School of Computer Science and Technology（计算机科学与技术学院） ； College of Computer Science and Artificial Intelligence（计算机科学与人工智能学院） ； Fudan University（复旦大学） ； Shanghai AI Laboratory（上海人工智能实验室） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.LG

AI总结 BioBridge结合蛋白质领域知识与通用语言能力，通过领域自适应持续预训练框架提升生物推理性能。

DermoGPT: Open Weights and Open Data for Morphology-Grounded Dermatological Reasoning MLLMs

DermoGPT: 开放权重和开放数据用于基于形态的皮肤病学推理大语言模型

Jinghan Ru, Siyuan Yan, Yuguo Yin, Yuexian Zou, Zongyuan Ge

机构 * School of Electronic and Computer Engineering, Peking University（电子与计算机工程学院，北京大学） ； Faculty of Information Technology, Monash University（信息科技学院，莫纳什大学） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.CL

AI总结 DermoGPT通过开放权重和数据提升皮肤病学推理的MLLM性能，实现与专家诊断流程的一致性。

LFC-DA: Logical Formula-Controlled Data Augmentation for Enhanced Logical Reasoning

Shenghao Li

机构 * Guangzhou University（广州大学） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.CL

Comments 10 pages, 6 figures

Large Language Models Imitate Logical Reasoning, but at what Cost?

Lachlan McGinness, Peter Baumgartner

机构 * Australian National University（澳大利亚国立大学） ； Data61|CSIRO 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.AI

Comments This work has been accepted as a main track paper for publication in the proceedings of the Australasian Joint Conference on Artificial Intelligence 2025 held in Canberra, Australia

Instantiation-based Formalization of Logical Reasoning Tasks using Language Models and Logical Solvers

Mohammad Raza, Natasa Milic-Frayling

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.AI

Comments IJCAI 2025

Building Trustworthy AI: Transparent AI Systems via Large Language Models, Ontologies, and Logical Reasoning (TranspNet)

Fadi Al Machot, Martin Thomas Horsch, Habib Ullah

机构 * Department of Data Science, Faculty of Science and Technology, Norwegian University of Life Sciences（数据科学系，科学与技术学院，挪威生命科学大学） 纠错

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.AI

Dialogue-based Explanations for Logical Reasoning using Structured Argumentation

Loan Ho, Stefan Schlobach

专题命中 逻辑推理 ：reasoning(title,abstract)；logical reasoning(title)；分类 cs.AI

Comments 45 pages, 8 gigures, journal