大模型推理能力 - arXivDaily 专题

2412.21187 2025-02-04 cs.CL 77%

Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs

Xingyu Chen, Jiahao Xu, Tian Liang, Zhiwei He, Jianhui Pang, Dian Yu, Linfeng Song, Qiuzhi Liu, Mengfei Zhou, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, Dong Yu

专题命中其他推理：reasoning(abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

Comments We have updated the results of DeepSeek-R1, and all conclusions still hold

2409.14306 2024-09-24 cs.AI 77%

LLMs are One-Shot URL Classifiers and Explainers

Fariza Rashid, Nishavi Ranaweera, Ben Doyle, Suranga Seneviratne

专题命中其他推理：reasoning(abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.AI

2404.15574 2024-04-25 cs.CL 77%

Retrieval Head Mechanistically Explains Long-Context Factuality

Wenhao Wu, Yizhong Wang, Guangxuan Xiao, Hao Peng, Yao Fu

专题命中其他推理：reasoning(abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

Comments Preprint

2312.17117 2023-12-29 cs.CV cs.AI 77%

Grounding-Prompter: Prompting LLM with Multimodal Information for Temporal Sentence Grounding in Long Videos

Houlun Chen, Xin Wang, Hong Chen, Zihan Song, Jia Jia, Wenwu Zhu

专题命中其他推理：reasoning(abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.AI

2311.09553 2023-11-17 cs.AI 77%

Program-Aided Reasoners (better) Know What They Know

Anubha Kabra, Sanketh Rangreji, Yash Mathur, Aman Madaan, Emmy Liu, Graham Neubig

专题命中其他推理：reasoning(abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.AI

2307.01640 2023-07-06 cs.CL 77%

Chain of Thought Prompting Elicits Knowledge Augmentation

Dingjun Wu, Jing Zhang, Xinmei Huang

专题命中其他推理：reasoning(abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

Comments Accepted by Findings of ACL 2023

2509.25100 2025-09-30 cs.LG cs.AI cs.CL 76%

ORPO-Distill: Mixed-Policy Preference Optimization for Cross-Architecture LLM Distillation

Aasheesh Singh, Vishal Vaddina, Dagnachew Birru

机构 * Phi Labs, Quantiphi（Phi Labs，Quantiphi）

专题命中其他推理：reasoning(abstract,comments)；CoT(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Accepted at NeurIPS 2025, Efficient Reasoning Workshop

2601.05451 2026-08-06 cs.LG cs.CL 版本更新 76%

RingSQL: Schema-Independent Synthetic Data Generation for Text-to-SQL Reinforcement Learning

RingSQL: 通过不依赖模式的模板生成合成数据以用于文本到SQL推理模型

Marko Sterbentz, Kevin Cushing, Cameron Barrie, Kristian J. Hammond

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.LG

AI总结 RingSQL通过结合不依赖模式的模板和大语言模型生成，提升文本到SQL模型的准确性和多样性

Comments 23 pages, 12 figures

2608.03161 2026-08-05 cs.AI cs.CL 新提交 76%

Evidence-Grounded Multimodal Knowledge Graph Construction for Multi-Lecture Educational Reasoning

面向多讲座教育推理的证据 grounded 多模态知识图谱构建

Sahil Al Farib, Momota Ahsana Meem, Sheikh Redwanul Islam, Md. Tanvir Raihan

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

AI总结该研究针对讲座视频知识保留不全问题，提出基于证据的多模态知识图谱构建流程，在神经网络讲座实验中取得高覆盖率与检索准确率，贡献可审计的知识图谱构建方法。

2607.20056 2026-07-23 cs.CL cs.AI 新提交 76%

Language-Specific versus Cross-Lingual Knowledge Graphs for Implicit Aspect Identification in Arabic: A Comparative Study of Reasoning and Adaptation Strategies

用于阿拉伯语隐式方面识别的特定语言与跨语言知识图谱：推理和适应策略的比较研究

Lujain A. Alawwad

机构 * Saudi Electronic University（沙特电子大学）

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

AI总结研究阿拉伯语隐式方面识别中特定语言与跨语言知识图谱策略，在混合管道中比较两种策略及生成提取器的零样本提示和特定任务微调两种适应策略，发现阿拉伯语本地KG效果更好，特定任务微调提升显著。

Comments 6 pages, 2 figures, 5 tables

2510.06843 2026-05-27 cs.CL cs.AI 76%

Self-signals Driven Multi-LLM Debate for Efficient and Accurate Reasoning

自信号驱动的多LLM辩论以实现高效准确的推理

Xuhang Chen, Zhifan Song, Deyi Ji, Shuo Gao, Lanyun Zhu

机构 * University of Cambridge（剑桥大学）； Sorbonne Université（索邦大学）； University of Science and Technology of China（中国科学技术大学）； Beihang University（北航大学）； Nanyang Technological University（南洋理工大学）

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

AI总结提出一种利用模型级置信度和token级语义焦点两种自信号来自适应引导多LLM辩论过程的方法，在提高准确性的同时减少token消耗。

2603.16880 2026-04-03 eess.SP cs.CL cs.LG q-bio.NC 76%

NeuroNarrator: A Generalist EEG-to-Text Foundation Model for Clinical Interpretation via Spectro-Spatial Grounding and Temporal State-Space Reasoning

NeuroNarrator：一种通过频谱-空间定位和时间状态空间推理的通用EEG到文本基础模型，用于临床解释

Guoan Wang, Shihao Yang, Jun-en Ding, Hao Zhu, Feng Liu

机构 * Stevens Institute of Technology（史蒂文斯理工学院）

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.LG

AI总结 NeuroNarrator通过频谱-空间定位和时间状态空间推理，将EEG信号转化为临床文本，解决传统EEG分析方法在临床解释中的局限性。

2603.05696 2026-03-09 cs.CE cs.AI cs.CL cs.NA math.NA 76%

Autonomous Algorithm Discovery for Ptychography via Evolutionary LLM Reasoning

基于进化大语言模型推理的自主算法发现用于ptychography

Xiangyu Yin, Ming Du, Junjing Deng, Zhi Yang, Yimo Han, Yi Jiang

机构 * Advanced Photon Source, Argonne National Laboratory, Lemont, IL, USA（阿贡国家实验室先进光子源）； Department of Materials Science and NanoEngineering, Rice University, Houston, TX, USA（材料科学与纳米工程系，德克萨斯大学里士满分校）

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

AI总结 Ptychi-Evolve通过进化大语言模型推理，自主发现并优化正则化算法，提升ptychography图像重建质量。

2603.01293 2026-03-03 cs.LG cs.AI stat.ML 76%

Theoretical Perspectives on Data Quality and Synergistic Effects in Pre- and Post-Training Reasoning Models

数据质量和协同效应在预训练和后训练推理模型中的理论视角

Adel Javanmard, Baharan Mirzasoleiman, Vahab Mirrokni

专题命中其他推理：reasoning(title)；分类 cs.AI、cs.LG

AI总结本文通过理论分析揭示了预训练和后训练模型中数据质量与协同效应的关键机制，发现平衡预训练数据能激活潜在能力，SFT在挑战性小样本上表现最佳，而RL在大规模非困难数据上更有效。

Comments 35 pages, 5 figures

2601.10274 2026-01-16 cs.LG cs.AI cs.IT cs.NI math.IT math.OC 76%

Queueing-Aware Optimization of Reasoning Tokens for Accuracy-Latency Trade-offs in LLM Servers

面向队列的推理令牌优化：在LLM服务器中准确率与延迟的权衡

Emre Ozbas, Melih Bastopcu

机构 * Department of Electrical and Electronics Engineering（电气与电子工程系）

专题命中其他推理：reasoning(title)；分类 cs.AI、cs.LG

AI总结本文提出了一种面向队列的推理令牌优化方法，通过优化LLM服务器中准确率与延迟的权衡，提升系统性能。

2512.12701 2025-12-16 cs.CV cs.CL cs.LG 76%

Efficient Vision-Language Reasoning via Adaptive Token Pruning

通过自适应令牌修剪实现高效的视觉语言推理

Xue Li, Xiaonan Song, Henry Hu

机构 * Scholar42（学者42）； InfiniPouch LLC（InfiniPouch公司）； Labelbox, Inc.（Labelbox公司）

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.LG

AI总结本研究提出自适应令牌修剪方法，通过动态保留关键令牌提升视觉语言模型的推理效率和稳定性，减少计算量并保持精度。

Comments 10 pages, 3 figures. Expanded version of an extended abstract accepted at NeurIPS 2025 Workshop on VLM4RWD. Presents methodology and preliminary experimental results

2509.10004 2025-09-15 cs.CL cs.AI 76%

Unsupervised Hallucination Detection by Inspecting Reasoning Processes

Ponhvoan Srey, Xiaobao Wu, Anh Tuan Luu

机构 * Nanyang Technological University（南洋理工大学）

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

Comments To appear in EMNLP 2025

2507.08235 2025-07-14 cs.LG cs.AI 76%

InsightBuild: LLM-Powered Causal Reasoning in Smart Building Systems

Pinaki Prasad Guha Neogi, Ahmad Mohammadshirazi, Rajiv Ramnath

机构 * Department of Computer Science（计算机科学系）； Engineering, Ohio State University, Ohio, US（工程学院，俄亥俄州立大学，俄亥俄州，美国）

专题命中其他推理：reasoning(title)；分类 cs.AI、cs.LG

2507.03327 2025-07-08 cs.CL cs.AI 76%

Read Quietly, Think Aloud: Decoupling Comprehension and Reasoning in LLMs

Yuanxin Wang, Ganesh Venkatesh

机构 * AppliedML

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

Comments Under submission

2503.10886 2025-03-17 cs.CV cs.AI cs.IR cs.LG q-bio.PE 76%

Taxonomic Reasoning for Rare Arthropods: Combining Dense Image Captioning and RAG for Interpretable Classification

Nathaniel Lesperance, Sujeevan Ratnasingham, Graham W. Taylor

专题命中其他推理：reasoning(title)；分类 cs.AI、cs.LG

Comments 12 pages, 3 figures

2501.08460 2025-01-16 cs.CV cs.AI cs.CL 76%

Towards Zero-Shot & Explainable Video Description by Reasoning over Graphs of Events in Space and Time

Mihai Masala, Marius Leordeanu

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

2412.11009 2024-12-17 cs.AI cs.CL cs.CY 76%

Dual Traits in Probabilistic Reasoning of Large Language Models

Shenxiong Li, Huaxia Rui

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

2410.14744 2024-10-22 cs.CL cs.AI 76%

Eliciting Uncertainty in Chain-of-Thought to Mitigate Bias against Forecasting Harmful User Behaviors

Anthony Sicilia, Malihe Alikhani

专题命中其他推理：chain-of-thought(title)；分类 cs.CL、cs.AI

2310.04562 2024-04-11 cs.CL cs.AI 76%

Towards Foundation Models for Knowledge Graph Reasoning

Mikhail Galkin, Xinyu Yuan, Hesham Mostafa, Jian Tang, Zhaocheng Zhu

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

Comments ICLR 2024

2402.15631 2024-02-27 cs.CL cs.AI 76%

Fine-Grained Self-Endorsement Improves Factuality and Reasoning

Ante Wang, Linfeng Song, Baolin Peng, Ye Tian, Lifeng Jin, Haitao Mi, Jinsong Su, Dong Yu

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.AI

2310.14435 2023-10-24 cs.CL cs.AI 76%

Retrieval-Augmented Chain-of-Thought in Semi-structured Domains

Vaibhav Mavi, Abulhair Saparov, Chen Zhao

专题命中其他推理：chain-of-thought(title)；分类 cs.CL、cs.AI

Comments to appear in NLLP 2023

2108.03834 2021-10-07 cs.AI cs.LG 76%

Bob and Alice Go to a Bar: Reasoning About Future With Probabilistic Programs

David Tolpin, Tomer Dobkin

专题命中其他推理：reasoning(title)；分类 cs.AI、cs.LG

Comments 31 pages, 9 figures, 2 tables

1804.05922 2018-04-18 cs.CL cs.LG 76%

Neural Models for Reasoning over Multiple Mentions using Coreference

Bhuwan Dhingra, Qiao Jin, Zhilin Yang, William W. Cohen, Ruslan Salakhutdinov

专题命中其他推理：reasoning(title)；分类 cs.CL、cs.LG

Comments NAACL 2018 (Short Paper)

2502.11521 2025-10-28 cs.CR cs.AI 76%

Detecting Various DeFi Price Manipulations with LLM Reasoning

Juantao Zhong, Daoyuan Wu, Ye Liu, Maoyi Xie, Yang Liu, Yi Li, Ning Liu

机构 * Daoyuan Wu is the co-first author.（达元吴是共同第一作者）； Ye Liu is the corresponding author.（叶刘是通讯作者）

专题命中其他推理：reasoning(title,comments)；分类 cs.AI

Comments Accepted by ASE 2025. Please cite the conference version of this paper, e.g., "Juantao Zhong, Daoyuan Wu, Ye Liu, Maoyi Xie, Yang Liu, Yi Li, Ning Liu. Detecting Various DeFi Price Manipulations with LLM Reasoning. In 40th IEEE/ACM International Conference on Automated Software Engineering (ASE 2025)"

2406.18930 2024-06-28 cs.AI cs.DM cs.LO cs.SC 76%

Reasoning About Action and Change

Florence Dupin de Saint-Cyr, Andreas Herzig, Jérôme Lang, Pierre Marquis

专题命中其他推理：reasoning(title,journal_ref)；分类 cs.AI

Journal ref Marquis, Pierre; Papini, Odile; Prade, Henri. A Guided Tour of Artificial Intelligence Research, 1 / 3, Springer International Publishing, pp.487-518, 2020, Knowledge Representation, Reasoning and Learning, 978-3-030-06163-0

1. 其他推理 5839 篇

Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs

LLMs are One-Shot URL Classifiers and Explainers

Retrieval Head Mechanistically Explains Long-Context Factuality

Grounding-Prompter: Prompting LLM with Multimodal Information for Temporal Sentence Grounding in Long Videos

Program-Aided Reasoners (better) Know What They Know

Chain of Thought Prompting Elicits Knowledge Augmentation

ORPO-Distill: Mixed-Policy Preference Optimization for Cross-Architecture LLM Distillation

RingSQL: Schema-Independent Synthetic Data Generation for Text-to-SQL Reinforcement Learning

Evidence-Grounded Multimodal Knowledge Graph Construction for Multi-Lecture Educational Reasoning

Language-Specific versus Cross-Lingual Knowledge Graphs for Implicit Aspect Identification in Arabic: A Comparative Study of Reasoning and Adaptation Strategies

Self-signals Driven Multi-LLM Debate for Efficient and Accurate Reasoning

NeuroNarrator: A Generalist EEG-to-Text Foundation Model for Clinical Interpretation via Spectro-Spatial Grounding and Temporal State-Space Reasoning

Autonomous Algorithm Discovery for Ptychography via Evolutionary LLM Reasoning

Theoretical Perspectives on Data Quality and Synergistic Effects in Pre- and Post-Training Reasoning Models

Queueing-Aware Optimization of Reasoning Tokens for Accuracy-Latency Trade-offs in LLM Servers

Efficient Vision-Language Reasoning via Adaptive Token Pruning

Unsupervised Hallucination Detection by Inspecting Reasoning Processes

InsightBuild: LLM-Powered Causal Reasoning in Smart Building Systems

Read Quietly, Think Aloud: Decoupling Comprehension and Reasoning in LLMs

Taxonomic Reasoning for Rare Arthropods: Combining Dense Image Captioning and RAG for Interpretable Classification

Towards Zero-Shot & Explainable Video Description by Reasoning over Graphs of Events in Space and Time

Dual Traits in Probabilistic Reasoning of Large Language Models

Eliciting Uncertainty in Chain-of-Thought to Mitigate Bias against Forecasting Harmful User Behaviors

Towards Foundation Models for Knowledge Graph Reasoning

Fine-Grained Self-Endorsement Improves Factuality and Reasoning

Retrieval-Augmented Chain-of-Thought in Semi-structured Domains

Bob and Alice Go to a Bar: Reasoning About Future With Probabilistic Programs

Neural Models for Reasoning over Multiple Mentions using Coreference

Detecting Various DeFi Price Manipulations with LLM Reasoning

Reasoning About Action and Change