大模型推理能力

Octopi: Object Property Reasoning with Large Tactile-Language Models

Samson Yu, Kelvin Lin, Anxing Xiao, Jiafei Duan, Harold Soh

专题命中 推理评测 ：reasoning(title,abstract)

Comments Accepted at Robotics: Science and Systems (R:SS 2024)

Momentor: Advancing Video Large Language Model with Fine-Grained Temporal Reasoning

Long Qian, Juncheng Li, Yu Wu, Yaobo Ye, Hao Fei, Tat-Seng Chua, Yueting Zhuang, Siliang Tang

专题命中 推理评测 ：reasoning(title,abstract)

Comments Accepted by ICML 2024

Retrieval Meets Reasoning: Even High-school Textbook Knowledge Benefits Multimodal Reasoning

Cheng Tan, Jingxuan Wei, Linzhuang Sun, Zhangyang Gao, Siyuan Li, Bihui Yu, Ruifeng Guo, Stan Z. Li

专题命中 推理评测 ：reasoning(title,abstract)

Comments Under review

How Good is my Video LMM? Complex Video Reasoning and Robustness Evaluation Suite for Video-LMMs

Muhammad Uzair Khattak, Muhammad Ferjad Naeem, Jameel Hassan, Muzammal Naseer, Federico Tombari, Fahad Shahbaz Khan, Salman Khan

专题命中 推理评测 ：reasoning(title,abstract)

Comments Technical report

LISA: Reasoning Segmentation via Large Language Model

Xin Lai, Zhuotao Tian, Yukang Chen, Yanwei Li, Yuhui Yuan, Shu Liu, Jiaya Jia

专题命中 推理评测 ：reasoning(title,abstract)

Comments Code, models, and data are available at https://github.com/dvlab-research/LISA

Demystifying Faulty Code with LLM: Step-by-Step Reasoning for Explainable Fault Localization

Ratnadira Widyasari, Jia Wei Ang, Truong Giang Nguyen, Neil Sharma, David Lo

专题命中 推理评测 ：reasoning(title,abstract)

Comments To be appeared at 2024 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER)

ViTCN: Vision Transformer Contrastive Network For Reasoning

Bo Song, Yuanhao Xu, Yichao Wu

专题命中 推理评测 ：reasoning(title,abstract)

Comments 5 pages, 2 figures , in proceeding of 5th International Seminar on Artificial Intelligence, Networking and Information Technology

Collaborative Position Reasoning Network for Referring Image Segmentation

Jianjian Cao, Beiya Dai, Yulin Li, Xiameng Qin, Jingdong Wang

专题命中 推理评测 ：reasoning(title,abstract)

Tracking with Human-Intent Reasoning

Jiawen Zhu, Zhi-Qi Cheng, Jun-Yan He, Chenyang Li, Bin Luo, Huchuan Lu, Yifeng Geng, Xuansong Xie

专题命中 推理评测 ：reasoning(title,abstract)

Comments 8 pages, 4 figures

Image Content Generation with Causal Reasoning

Xiaochuan Li, Baoyu Fan, Runze Zhang, Liang Jin, Di Wang, Zhenhua Guo, Yaqian Zhao, Rengang Li

专题命中 推理评测 ：reasoning(title,abstract)

Comments Accepted by the 38th Annual AAAI Conference on Artificial Intelligence (AAAI 2024) in December 2023

Lenna: Language Enhanced Reasoning Detection Assistant

Fei Wei, Xinyu Zhang, Ailing Zhang, Bo Zhang, Xiangxiang Chu

专题命中 推理评测 ：reasoning(title,abstract)

Demo2Code: From Summarizing Demonstrations to Synthesizing Code via Extended Chain-of-Thought

Huaxiaoyue Wang, Gonzalo Gonzalez-Pumariega, Yash Sharma, Sanjiban Choudhury

专题命中 推理评测 ：chain-of-thought(title,abstract)

Comments 10 pages (not including references and appendix), 14 figures (7 in main paper, 7 in appendix); (v3) camera-ready version

RoboVQA: Multimodal Long-Horizon Reasoning for Robotics

Pierre Sermanet, Tianli Ding, Jeffrey Zhao, Fei Xia, Debidatta Dwibedi, Keerthana Gopalakrishnan, Christine Chan, Gabriel Dulac-Arnold, Sharath Maddineni, Nikhil J Joshi, Pete Florence, Wei Han, Robert Baruch, Yao Lu, Suvir Mirchandani, Peng Xu, Pannag Sanketi, Karol Hausman, Izhak Shafran, Brian Ichter, Yuan Cao

专题命中 推理评测 ：reasoning(title,abstract)

Benchmarking Sequential Visual Input Reasoning and Prediction in Multimodal Large Language Models

Mingwei Zhu, Leigang Sha, Yu Shu, Kangjia Zhao, Tiancheng Zhao, Jianwei Yin

专题命中 推理评测 ：reasoning(title,abstract)

CoT-MISR:Marrying Convolution and Transformer for Multi-Image Super-Resolution

Mingming Xiu, Yang Nie, Qing Song, Chun Liu

专题命中 推理评测 ：CoT(title,abstract)

VTQA: Visual Text Question Answering via Entity Alignment and Cross-Media Reasoning

Kang Chen, Xiangqian Wu

专题命中 推理评测 ：reasoning(title,abstract)

Learning to Agree on Vision Attention for Visual Commonsense Reasoning

Zhenyang Li, Yangyang Guo, Kejie Wang, Fan Liu, Liqiang Nie, Mohan Kankanhalli

专题命中 推理评测 ：reasoning(title,abstract)

Parallel Reasoning Network for Human-Object Interaction Detection

Huan Peng, Fenggang Liu, Yangguang Li, Bin Huang, Jing Shao, Nong Sang, Changxin Gao

专题命中 推理评测 ：reasoning(title,abstract)

Precondition and Effect Reasoning for Action Recognition

Yoo Hongsang, Li Haopeng, Ke Qiuhong, Liu Liangchen, Zhang Rui

专题命中 推理评测 ：reasoning(title,abstract)

Comments The paper is under consideration at Computer Vision and Image Understanding

HERO: HiErarchical spatio-tempoRal reasOning with Contrastive Action Correspondence for End-to-End Video Object Grounding

Mengze Li, Tianbao Wang, Haoyu Zhang, Shengyu Zhang, Zhou Zhao, Wenqiao Zhang, Jiaxu Miao, Shiliang Pu, Fei Wu

专题命中 推理评测 ：reasoning(title,abstract)

Visual Abductive Reasoning

Chen Liang, Wenguan Wang, Tianfei Zhou, Yi Yang

专题命中 推理评测 ：reasoning(title,abstract)

Comments CVPR2022; Code, data: https://github.com/leonnnop/VAR

Improving Object Permanence using Agent Actions and Reasoning

Ying Siu Liang, Chen Zhang, Dongkyu Choi, Kenneth Kwok

专题命中 推理评测 ：reasoning(title,abstract)

Comments 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2021)

How Transferable are Reasoning Patterns in VQA?

Corentin Kervadec, Theo Jaunet, Grigory Antipov, Moez Baccouche, Romain Vuillemot, Christian Wolf

专题命中 推理评测 ：reasoning(title,abstract)

Transformer Reasoning Network for Image-Text Matching and Retrieval

Nicola Messina, Fabrizio Falchi, Andrea Esuli, Giuseppe Amato

专题命中 推理评测 ：reasoning(title,abstract)

Comments Presented at ICPR 2020

Do AI Agents Know When a Task Is Simple? Toward Complexity-Aware Reasoning and Execution

人工智能代理知道任务何时简单吗？迈向复杂性感知推理与执行

Junjie Yin, Xinyu Feng

专题命中 推理评测 ：reasoning(title,comments)；分类 cs.CL、cs.AI

AI总结 研究探讨LLM代理缺乏任务感知执行范围估计能力，提出E3方法，在MSE-Bench基准测试及真实模型工具中验证，该方法能在保证成功率时大幅降低成本，推动迈向工程基础人工智能，还发布了框架和基准。

Comments 27 pages, 8 figures, 8 tables. Code and benchmark: https://github.com/eejyin/Do-AI-Agents-Know-When-a-Task-Is-Simple-Toward-Complexity-Aware-Reasoning-and-Execution

LAUDE: LLM-Assisted Unit Test Generation and Debugging of Hardware DEsigns

LAUDE: 基于LLM的硬件设计单元测试生成与调试

Deeksha Nandal, Riccardo Revalor, Soham Dan, Debjit Pal

机构 * Dept. of Electrical and Computer Engineering, University of Illinois Chicago（伊利诺伊大学芝加哥分校电子与计算机工程系） ； Microsoft（微软公司） 纠错

专题命中 推理评测 ：reasoning(abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.AI

AI总结 LAUDE利用大语言模型能力，实现硬件设计的单元测试生成与调试，有效检测和修复设计中的错误。

Comments 11 Pages, 9 Figures, 9 Tables Submitted to AAAI 2027

Gaokerena: A Small Persian Medical Language Model Family

Gaokerena：一个小型波斯语医疗语言模型家族

Mehrdad Ghassabi, Hamidreza Baradaran Kashani, Pedram Rostami, Sadra Hakim, Zahra Kazemi, Audrina Ebrahimi

机构 * University of Isfahan（伊斯法罕大学） ； University of Tehran（德黑兰大学） ； University of Windsor（温莎大学） ； Alzahra University（阿勒扎哈拉大学） ； University of Texas at Dallas（德克萨斯大学达拉斯分校） 纠错

专题命中 推理评测 ：chain-of-thought(abstract,abstract_cn)；reasoning(abstract)；分类 cs.CL

AI总结 针对波斯语医疗语言模型研究不足的问题，该研究推出Gaokerena家族，包含Gaokerena-V和Gaokerena-R两款模型，在医疗问答基准上取得性能提升，还配备不确定性头，为本地化数字医疗提供基础但仍需完善。

Comments 29 pages, 9 figures

To Compare, or Not to Compare: On Methodological Practices in Evaluating Social Bias

比较还是不比较：论评估社会偏见的方法论实践

Federico Marcuzzi, Xuefei Ning, Roy Schwartz, Iryna Gurevych

机构 * Tsinghua University（清华大学） ； The Hebrew University of Jerusalem（耶路撒冷希伯来大学） ； Ubiquitous Knowledge Processing Lab (UKP Lab), Department of Computer Science, TU Darmstadt and National Research Center for Applied Cybersecurity ATHENE（普适知识处理实验室（UKP Lab），计算机科学系，达姆施塔特工业大学及国家应用网络安全研究中心ATHENE） 纠错

专题命中 推理评测 ：reasoning(abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

AI总结 提出统一框架标准化异构基准，揭示孤立评估与强制比较设置间的系统性范式差距，发现比较设置会催化潜在歧视，且链式思维推理加剧偏见，规模越大偏见越强。

ThinkDeception: A Progressive Reinforcement Learning Framework for Interpretable Multimodal Deception Detection

ThinkDeception: 一种用于可解释多模态欺骗检测的渐进式强化学习框架

Jinhao Song, Shan Liang, Yiqun Yue, Zhuhuayang Zhang, Tianqi Gao

机构 * Xi'an Jiaotong-Liverpool University（西安交通大学利物浦大学） 纠错

专题命中 推理评测 ：CoT(abstract,abstract_cn)；reasoning(abstract)；分类 cs.AI

AI总结 提出ThinkDeception框架，将多模态大语言模型引入欺骗检测，通过逐步推理和视觉-音频一致性组相对策略优化（VAC-GRPO）实现可解释的认知推理，在主流基准上达到新SOTA。

Comments 10pages,4figures

Simulating Students' Java Programming Errors with Large Language Models

用大型语言模型模拟学生的Java编程错误

Ali Keramati, Jie Cao, Iman Mohammadi, Mark Warschauer, Yang Shi

机构 * University of California, Los Angeles（加州大学洛杉矶分校） 纠错

专题命中 推理评测 ：CoT(abstract,abstract_cn)；chain-of-thought(abstract)；分类 cs.CL

AI总结 探索用LLM模拟学生编程错误，评估五种模型在多样性和对齐性上的表现，发现Claude Sonnet 4平衡最佳，且合成错误与真实错误难以区分。