Comments2: corrected NLP-mode precision/recall after a recogniser fix that lifts the US_SSN and PHONE detection v1 understated, and revised the recommended threshold to min_score=0.5 (the curve is flat, not peaked). Section 5 tables, the threshold analysis, and Figures 1-2 are updated; headline micro-F1 = 0.898
Beyond Parallel Sampling: Diverse Query Initialization for Agentic Search
超越并行采样:面向智能搜索的多样化查询初始化
Sidhaarth Murali, João Coelho, Jingjie Ning, João Magalhães, Bruno Martins, Chenyan Xiong
机构
*
Carnegie Mellon University(卡内基梅隆大学)
;
Instituto Superior Técnico and INESC-ID, University of Lisbon(里斯本大学高等技术学院和INESC-ID)
;
NOVA LINCS, NOVA School of Science and Technology(新里斯本大学科学与技术学院NOVA LINCS)
Parallelizing Tool Execution and LLM Generation for Low-Latency Agent Serving
并行化工具执行与LLM生成以实现低延迟代理服务
Yifan Sui, Han Zhao, Rui Ma, Zhiyuan He, Hao Wang, Jianxun Li, Kaiqiang Xu, Kai Chen, Yuqing Yang
机构
*
Shanghai Jiao Tong University(上海交通大学)
;
Microsoft Research(微软研究院)
;
Stevens Institute of Technology(Stevens 工程学院)
;
Google(谷歌)
;
Hong Kong University of Science and Technology(香港科学与技术大学)
Visual-Seeker: Towards Visual-Native Multimodal Agentic Search via Active Visual Reasoning
Visual-Seeker:通过主动视觉推理实现视觉原生多模态智能搜索
Zhengbo Zhang, Changtao Miao, Jinbo Su, Zhaowen Zhou, Chunxia Zhang, Xukai Wang, Ruiqi Liu, Kaiyuan Zheng, Jiansheng Cai, Bo Zhang, Zhe Li, Shiming Xiang, Ying Yan
机构
*
School of Artificial Intelligence UCAS(中国科学院大学人工智能学院)
;
Institute of Automation CAS(中国科学院自动化研究所)
;
Ant Digital Technologies Ant Group(蚂蚁数字科技蚂蚁集团)
;
RUC(中国人民大学)
;
BIT(北京理工大学)
R2IF: Aligning Reasoning with Decisions via Composite Rewards for Interpretable LLM Function Calling
R2IF: 通过复合奖励对齐推理与决策以实现可解释的LLM函数调用
Aijia Cheng, Kailong Wang, Ling Shi, Yongxin Zhao
机构
*
Shanghai Key Laboratory of Trustworthy Computing, East China Normal University, Shanghai, China(上海可信计算实验室,华东师范大学,上海,中国)
;
Huazhong University of Science and Technology(华中科技大学)
;
Nanyang Technological University(南洋理工大学)