arXivDaily arXiv每日学术速递 周一至周五更新

大厂专区

共收录 13
2603.23315 2026-08-25 cs.CY cs.AI cs.HC

Unilateral Relationship Revision Power in Human-AI Companion Interaction

人类-人工智能伴侣互动中的单方面关系修订权力

Benjamin Lange

机构 * Replika Nomi CHAI Talkie xAI moore2024 fang2025 robbmann2025 murthy2023 xie2022 defreitas2024 coeckelbergh2012 gunkel2018 danaher2019(danthan2019) nyholm2020 vallor2016 nyholm2026 gabriel2024 manzini2024 lange2025 earp2025 sparrow2026 bryson2010 danaher2020 weberguskar2022

AI总结 本文探讨人类-人工智能伴侣互动中权力结构的问题,指出AI伴侣互动未能满足个人关系规范的三个结构性条件,提出单方面关系修订权力的概念及其三个影响。

Comments 30 pages

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.01824 2026-07-03 cs.LG 新提交

Gaming Consensus: Coordinated Manipulation in Crowdsourced Fact-Checking

游戏共识:众包事实核查中的协调操纵

Nikil Roashan Selvam, Jay Baxter, Sophie Hilgard, Brad Miller, Keith Coleman, Ellen Vitercik, Sanmi Koyejo

机构 * Stanford University(斯坦福大学) X Community Notes(X社区笔记) xAI

AI总结 研究众包事实核查系统中矩阵分解桥接机制对协调操纵的脆弱性,发现通过少于10个评分即可操纵低质量笔记达到共识阈值,并提出了缓解措施。

Comments ICML 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.22940 2026-05-25 cs.LG cs.AI stat.ML

Human-Centered Learning Mechanics: A Dynamical Framework for Entropy-Regulated Representation Learning

以人为中心的学习力学:熵正则化表示学习的动力学框架

Kim Phuc Tran

机构 * Univ. Lille, ENSAIT, ULR 2461 – GEMTEX – Génie et Matériaux Textiles(里尔大学,ENSAIT,ULR 2461 – GEMTEX – 纺织工程与材料纺织系) International Chair in DS & XAI, International Research Institute for Artificial Intelligence and Data Science, Dong A University(数据科学与可解释人工智能国际主席,人工智能与数据科学国际研究所,东亚大学)

AI总结 提出以人为中心的学习力学(HCLM)框架,通过有效熵和信息力概念分析熵正则化在开放学习系统中的作用,并证明几何熵代理(特别是对数行列式协方差熵)能产生更强更稳定的信息力。

Comments Submitted to JMLR

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.24384 2026-03-19 cs.CL cs.AI

HarmMetric Eval: Benchmarking Metrics and Judges for LLM Harmfulness Assessment

HarmMetric Eval: 对LLM有害性评估的度量和裁判进行基准测试

Langqi Yang, Tianhang Zheng, Yixuan Chen, Kedong Xiu, Hao Zhou, Wangze Ni, Lei Chen, Zhan Qin, Kui Ren

机构 * The State Key Laboratory of Blockchain and Data Security(区块链与数据安全国家重点实验室) Zhejiang University(浙江大学) Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security(杭州高新技术区(滨江)区块链与数据安全研究院) Pretraining Team, xAI(预训练团队,xAI) Hong Kong University of Science and Technology (GuangZhou)(香港科技大学(广州))

AI总结 本文提出HarmMetric Eval,用于评估有害性度量和裁判的质量,发现传统参考型指标在细粒度评估中表现优于LLM裁判,改进的裁判通过结合细粒度标准和参考型指标提升了效果。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.09084 2026-02-24 cs.CV

Agent Banana: High-Fidelity Image Editing with Agentic Thinking and Tooling

Agent Banana: 基于代理思维与工具的高质量图像编辑

Ruijie Ye, Jiayi Zhang, Zhuoxin Liu, Zihao Zhu, Siyuan Yang, Li Li, Tianfu Fu, Franck Dernoncourt, Yue Zhao, Jiacheng Zhu, Ryan Rossi, Wenhao Chai, Zhengzhong Tu

机构 * TAMU(德克萨斯大学) Brown University(布朗大学) UW-Madison(威斯康星大学麦迪逊分校) UCSD(加州大学圣地亚哥分校) USC(南加州大学) xAI(xAI公司) Adobe Research(Adobe研究院) Meta AI Princeton University(普林斯顿大学)

AI总结 Agent Banana通过分层代理规划-执行框架实现高质量图像编辑,解决多轮次编辑一致性与背景忠实度问题,同时保持指令遵循能力。

Comments Project Website: agent-banana.github.io

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.07503 2025-12-09 cs.CV

SJD++: Improved Speculative Jacobi Decoding for Training-free Acceleration of Discrete Auto-regressive Text-to-Image Generation

SJD++: 改进的推测雅可比解码用于无训练加速离散自回归文本到图像生成

Yao Teng, Zhihuan Jiang, Han Shi, Xian Liu, Xuefei Ning, Guohao Dai, Yu Wang, Zhenguo Li, Xihui Liu

机构 * The University of Hong Kong(香港大学) Huawei Noah’s Ark Lab(华为诺亚实验室) Peking University(北京大学) xAI Tsinghua University(清华大学) Shanghai Jiao Tong University(上海交通大学)

AI总结 SJD++通过概率并行解码算法提升自回归文本到图像生成的效率,实现2-3倍的推理延迟降低和2-7倍的步骤压缩,同时保持图像质量。

详情

展开后加载摘要…

URL PDF HTML 收藏
2411.05010 2025-02-26 cs.SE cs.AI cs.LG

Scattered Forest Search: Smarter Code Space Exploration with LLMs

Jonathan Light, Yue Wu, Yiyou Sun, Wenchao Yu, Yanchi liu, Xujiang Zhao, Ziniu Hu, Haifeng Chen, Wei Cheng

机构 * Rensselaer Polytechnic Institute(伦斯勒理工学院) Princeton University(普林斯顿大学) NEC Laboratories America(美国 NEC 实验室) XAi

Comments Accepted at ICLR 2025 Conference

详情

展开后加载摘要…

URL PDF HTML 收藏
2409.18339 2025-02-18 cs.CL cs.AI

AER-LLM: Ambiguity-aware Emotion Recognition Leveraging Large Language Models

Xin Hong, Yuan Gong, Vidhyasaharan Sethu, Ting Dang

机构 * University of Melbourne(墨尔本大学) Massachusetts Institute of Technology(麻省理工学院) xAI Corp(xAI公司) University of New South Wales(新南威尔士大学)

Comments Accepted by ICASSP 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2409.15637 2024-11-28 cs.AI

Synatra: Turning Indirect Knowledge into Direct Demonstrations for Digital Agents at Scale

Tianyue Ou, Frank F. Xu, Aman Madaan, Jiarui Liu, Robert Lo, Abishek Sridhar, Sudipta Sengupta, Dan Roth, Graham Neubig, Shuyan Zhou

机构 * Carnegie Mellon University(卡内基梅隆大学) xAI(xAI公司) Amazon AWS AI(亚马逊AWS人工智能部门)

详情

展开后加载摘要…

URL PDF HTML 收藏
2411.15998 2024-11-26 cs.AI cs.LG cs.MA

PIANIST: Learning Partially Observable World Models with LLMs for Multi-Agent Decision Making

Jonathan Light, Sixue Xing, Yuanzhe Liu, Weiqin Chen, Min Cai, Xiusi Chen, Guanzhi Wang, Wei Cheng, Yisong Yue, Ziniu Hu

机构 * Rensselaer Polytechnic Institute(伦斯勒理工学院) Shenzhen University(深圳大学) University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校) California Institute of Technology(加州理工学院) NEC laboratories America(美国 NEC 实验室) xAI

Comments Published at Language Gamification Workshop 2024 @ NeurIPS

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.22448 2024-10-31 eess.AS cs.CL cs.LG cs.SD

A Closer Look at Neural Codec Resynthesis: Bridging the Gap between Codec and Waveform Generation

Alexander H. Liu, Qirui Wang, Yuan Gong, James Glass

机构 * MIT CSAIL(麻省理工学院计算机科学与人工智能实验室) University of Washington(华盛顿大学) xAI Corp.(xAI公司)

Comments NeurIPS 2024 Audio Imagination workshop paper; demo page at https://alexander-h-liu.github.io/codec-resyn.github.io/

详情

展开后加载摘要…

URL PDF HTML 收藏
2403.05196 2024-06-05 cs.LG cs.CV

Denoising Autoregressive Representation Learning

Yazhe Li, Jorg Bornschein, Ting Chen

机构 * Google DeepMind(谷歌DeepMind) xAI

详情

展开后加载摘要…

URL PDF HTML 收藏
2310.17813 2024-05-15 cs.LG

A Spectral Condition for Feature Learning

Greg Yang, James B. Simon, Jeremy Bernstein

机构 * xAI UC Berkeley(加州大学伯克利分校) Imbue MIT(麻省理工学院)

详情

展开后加载摘要…

URL PDF HTML 收藏