arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~

AI 大模型

大模型推理能力

大模型数学、逻辑、规划、多步推理和测试时计算能力。

2026-02-23 至 2026-02-23 共收录 8 信号源:cs.CL, cs.AI, cs.LG

1. 推理评测 8 篇

2601.19318 2026-02-23 cs.RO cs.CV 82%

Perception-to-Pursuit: Track-Centric Temporal Reasoning for Open-World Drone Detection and Autonomous Chasing

感知到追捕:面向开放世界的无人机检测与自主追捕的以轨迹为中心的时序推理

Venkatakrishna Reddy Oruganti

机构 * Sithara Inc.(Sithara公司)

专题命中 推理评测 :reasoning(title,abstract);planning(abstract)

AI总结 P2P通过轨迹为中心的时序推理框架,提升了无人机轨迹预测精度和追捕可行性,实现了准确的预测与可操作的追捕规划。

Comments 7 pages, 2 figures, 3 tables, 15 references. Intended for submission to ICCV 2027

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.17592 2026-02-23 astro-ph.IM cs.LG 79%

AstroMLab 4: Benchmark-Topping Performance in Astronomy Q&A with a 70B-Parameter Domain-Specialized Reasoning Model

AstroMLab 4: 在天文学问答中通过700亿参数领域专用模型实现基准顶级性能

Tijmen de Haan, Yuan-Sen Ting, Tirthankar Ghosal, Tuan Dung Nguyen, Alberto Accomazzi, Emily Herron, Vanessa Lama, Rui Pan, Azton Wells, Nesar Ramachandra

机构 * Institute of Particle Nuclear Studies (IPNS), High Energy Accelerator Research Organization (KEK), Tsukuba, Ibaraki 305-0801, Japan International Center for Quantum-field Measurement Systems for Studies of the Universe Particles (QUP-WPI), High Energy Accelerator Research Organization (KEK), Tsukuba, Ibaraki 305-0801, Japan Department of Astronomy, The Ohio State University, Columbus, OH, USA Center for Cosmology AstroParticle Physics (CCAPP), The Ohio State University, Columbus, OH, USA National Center for Computational Sciences, Oak Ridge National Laboratory, Oak Ridge, TN, USA Department of Computer Information Science, University of Pennsylvania, Philadelphia, PA, USA Center for Astrophysics, Harvard \& Smithsonian, Cambridge, MA, USA Siebel School of Computing Data Science, University of Illinois at Urbana-Champaign, Urbana-Champaign, IL, USA Computational Science Division, Argonne National Laboratory, Lemont, IL, USA

专题命中 推理评测 :reasoning(title,abstract);分类 cs.LG

AI总结 AstroSage-Llama-3.1-70B通过700亿参数领域专用模型在天文学问答中实现顶级性能,优于GPT-5.2等通用模型。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.18137 2026-02-23 cs.CL cs.AI cs.LG 67%

Agentic Adversarial QA for Improving Domain-Specific LLMs

代理对抗问答:提升领域特定大语言模型

Vincent Grari, Ciprian Tomoiaga, Sylvain Lamprier, Tatsunori Hashimoto, Marcin Detyniecki

机构 * Stanford University(斯坦福大学) Polish Academy of Science, IBS PAN, Warsaw, Poland(波兰科学院、IBS PAN、华沙、波兰)

专题命中 推理评测 :reasoning(abstract);分类 cs.CL、cs.AI、cs.LG

AI总结 本研究提出了一种对抗性问题生成框架,通过生成语义具有挑战性的问题来提升领域特定大语言模型的适应能力。

Comments 9 pages, 1 Figure

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.17689 2026-02-23 cs.LG cs.AI cs.CL cs.CV 67%

Robust Pre-Training of Medical Vision-and-Language Models with Domain-Invariant Multi-Modal Masked Reconstruction

具有领域不变多模态掩码重建的医学视觉-语言模型鲁棒预训练

Melika Filvantorkaman, Mohsen Piri

专题命中 推理评测 :reasoning(abstract);分类 cs.CL、cs.AI、cs.LG

AI总结 本文提出Robust-MMR框架,通过显式建模鲁棒性提升医学视觉-语言模型在跨领域和扰动下的表现,实现更可靠的医疗应用。

Comments 28 pages, 3 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.17672 2026-02-23 cs.HC cs.AI cs.CL cs.CR cs.CY 62%

Assessing LLM Response Quality in the Context of Technology-Facilitated Abuse

评估在技术辅助虐待情境下的LLM响应质量

Vijay Prakash, Majed Almansoori, Donghan Hu, Rahul Chatterjee, Danny Yuxing Huang

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

AI总结 本文评估了四种LLM在技术辅助虐待情境下的响应质量,揭示其在支持幸存者方面的有效性和局限性,并提出改进建议。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.15854 2026-02-23 cs.CL cs.AI 62%

Decoupling Strategy and Execution in Task-Focused Dialogue via Goal-Oriented Preference Optimization

通过目标导向的偏好优化实现任务导向对话中的解耦策略与执行

Jingyi Xu, Xingyu Ren, Zhoupeng Shou, Yumeng Zhang, Zhiqiang You

机构 * School of Information Engineering, China Jiliang University, Hangzhou, China(信息工程学院,中国浙江大学,杭州,中国) College of Chemical and Biological Engineering, Zhejiang University, Hangzhou, China(化学与生物工程学院,浙江大学,杭州,中国)

专题命中 推理评测 :planning(abstract);分类 cs.CL、cs.AI

AI总结 本文提出GOPO框架,通过解耦策略规划与响应生成,提升任务导向对话系统的长周期任务成功率。

详情

展开后加载摘要…

URL PDF HTML 收藏
2504.17311 2026-02-23 cs.CL cs.AI 62%

FLUKE: A Linguistically-Driven and Task-Agnostic Framework for Robustness Evaluation

FLUKE:一种基于语言驱动且任务无关的鲁棒性评估框架

Yulia Otmakhova, Hung Thinh Truong, Rahmad Mahendra, Zenan Zhai, Rongxin Zhu, Daniel Beck, Jey Han Lau

机构 * The University of Melbourne(墨尔本大学) Oracle(Oracle公司) Universitas Indonesia(印度尼西亚大学) RMIT University(皇家墨尔本理工学院)

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

AI总结 FLUKE通过系统性语言变化评估模型鲁棒性,揭示任务依赖性、模型脆弱性和语言特征使用与鲁棒性无关的结论。

Comments Accepted to EACL 2026 Findings

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.13662 2026-02-23 cs.CV cs.AI 57%

LeafNet: A Large-Scale Dataset and Comprehensive Benchmark for Foundational Vision-Language Understanding of Plant Diseases

LeafNet:一个大规模数据集和全面基准,用于植物病害的基础视觉-语言理解

Khang Nguyen Quoc, Phuong D. Dao, Luyl-Da Quach

机构 * School of Electrical Engineering, Korea University(韩国大学电气工程学院) Department of Integrative Biology, The University of Texas at Austin(德克萨斯大学奥斯汀分校整合生物学系) Department of Software Engineering, FPT University(FPT大学软件工程系)

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

AI总结 LeafNet通过大规模多模态数据集和基准测试,揭示了VLMs在植物疾病理解中的性能差异,强调了多模态架构在提升诊断精度中的关键作用。

Comments 26 pages, 13 figures and 8 tables

详情

展开后加载摘要…

URL PDF HTML 收藏