arXivDaily arXiv每日学术速递 周一至周五更新

大厂专区

Google(谷歌)

2026-03-18 至 2026-03-18 共收录 6
2603.16864 2026-03-18 cs.CV cs.AI

SparkVSR: Interactive Video Super-Resolution via Sparse Keyframe Propagation

SparkVSR:通过稀疏关键帧传播实现交互式视频超分辨率

Jiongze Yu, Xiangbo Gao, Pooja Verlani, Akshay Gadde, Yilin Wang, Balu Adsumilli, Zhengzhong Tu

机构 * Texas A\&M University(德克萨斯A&M大学) YouTube, Google(YouTube和谷歌)

AI总结 SparkVSR提出一种交互式视频超分辨率框架,通过稀疏关键帧作为控制信号,提升时间一致性和恢复质量,实验显示在多个基准上优于基线方法。

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.15968 2026-03-18 cs.AI cs.CL cs.LG cs.MA

MAC: Multi-Agent Constitution Learning

MAC:多智能体宪法学习

Rushil Thareja, Gautam Gupta, Francesco Pinto, Nils Lukas

机构 * Mohamed bin Zayed University of Artificial Intelligence(莫扎德人工智能大学) Indraprastha Institute of Information Technology Delhi(德里印度教教派信息科技学院) Google DeepMind(谷歌DeepMind)

AI总结 本文提出MAC多智能体宪法学习方法,通过结构化提示优化提升LLM控制效果,实现可解释的规则集生成,优于现有提示优化方法,并在PII标注任务中表现优异。

Comments Code: https://github.com/rushil-thareja/MAC-Multi-Agent-Constitution-Learning | PyPI: https://pypi.org/project/mac-prompt/ | Website: https://www.mac-prompt.com/

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.15563 2026-03-18 cs.LG cs.AI

The PokeAgent Challenge: Competitive and Long-Context Learning at Scale

PokeAgent挑战:在大规模中实现竞争性和长上下文学习

Seth Karten, Jake Grigsby, Tersoo Upaa, Junik Bae, Seonghun Hong, Hyunyoung Jeong, Jaeyoon Jung, Kun Kerdthaisong, Gyungbo Kim, Hyeokgi Kim, Yujin Kim, Eunju Kwon, Dongyu Liu, Patrick Mariglia, Sangyeon Park, Benedikt Schink, Xianwei Shi, Anthony Sistilli, Joseph Twin, Arian Urdu, Matin Urdu, Qiao Wang, Ling Wu, Wenli Zhang, Kunsheng Zhou, Stephanie Milani, Kiran Vodrahalli, Amy Zhang, Fei Fang, Yuke Zhu, Chi Jin

机构 * Princeton(普林斯顿大学) UT-Austin(得克萨斯大学奥斯汀分校) CMU(卡内基梅隆大学) NYU(纽约大学) Google DeepMind(谷歌DeepMind) Team Heatz(团队Heatz) Team PA-Agent(团队PA-Agent) Team FoulPlay(团队FoulPlay) Team 4thLesson(团队4thLesson) Team Q(团队Q) Team Anthonys(团队Anthonys) Team Hamburg(团队Hamburg) Team Porygon2AI(团队Porygon2AI) Team Deepest(团队Deepest) Team August(团队August)

AI总结 PokeAgent挑战通过两个互补赛道,解决部分可观测性、博弈推理和长周期规划问题,提供大规模基准测试和首个RPG速run评估框架,揭示通用(LLM)、专业(RL)和精英人类表现间的差距。

Comments 41 pages, 26 figures, 5 tables. NeurIPS 2025 Competition Track

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.25140 2026-03-18 cs.AI cs.CL

ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory

ReasoningBank: 通过推理记忆扩展智能体自我进化

Siru Ouyang, Jun Yan, I-Hung Hsu, Yanfei Chen, Ke Jiang, Zifeng Wang, Rujun Han, Long T. Le, Samira Daruki, Xiangru Tang, Vishy Tirumalashetty, George Lee, Mahsan Rofouei, Hangfei Lin, Jiawei Han, Chen-Yu Lee, Tomas Pfister

机构 * University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校) Google Cloud AI Research(谷歌云人工智能研究) Yale University(耶鲁大学) Google Cloud AI(谷歌云人工智能)

AI总结 本文提出ReasoningBank,通过整合智能体自我评估的成功与失败经验,提升其持续学习能力,并引入MaTTS加速学习过程,提升任务表现与效率。

Comments Accepted to ICLR 2026; Code: https://github.com/google-research/reasoning-bank

详情

展开后加载摘要…

URL PDF HTML 收藏
2405.15216 2026-03-18 cs.LG cs.CL cs.SD eess.AS

Revisiting ASR Error Correction with Specialized Models

重新审视基于专用模型的语音识别纠错

Zijin Gu, Tatiana Likhomanenko, He Bai, Erik McDermott, Ronan Collobert, Navdeep Jaitly

机构 * Apple(苹果公司) Google(谷歌)

AI总结 本文提出基于紧凑seq2seq模型的语音识别纠错方法,通过合成语料训练,实现更高效的纠错性能,优于大型语言模型并在多种ASR架构和领域中表现优异。

Comments under review

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.15811 2026-03-18 cs.CV

Feed-forward Gaussian Registration for Head Avatar Creation and Editing

前馈高斯注册用于头像创建与编辑

Malte Prinzler, Paulo Gotardo, Siyu Tang, Timo Bolkart

机构 * Google(谷歌)

AI总结 MATCH方法通过高效高斯注册实现高质量头像创建,无需预处理,速度快且支持跨主体应用,优于现有方法。

Comments Website: https://malteprinzler.github.io/projects/match ; Video: https://youtu.be/Z3xoXQ648sE

详情

展开后加载摘要…

URL PDF HTML 收藏