paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达
登录 DINQ
返回新星榜
CY

Cheng Yang

ACL 2026×3ICLR 2026智能体与工具推理评测与基准代码强化学习与后训练
新星分新星
72
一作贡献30
核心/参与15
方向广度13
合作网络12
1 一作 · 4 篇 · ACL 2026 / ICLR 2026

获取完整研究档案

首次获取该候选人的完整档案及 StartLux 评估共 20 积分;每位用户单独解锁,之后查看免费。主动刷新 20 积分。

联系方式

实时查找电子邮箱,费用不超过20积分;未找到邮箱不扣费。

2026 顶会论文

4
一作ICLR 2026Poster· Reinforcement Learning / Multi-agent

From What to Why: A Multi-Agent System for Evidence-based Chemical Reaction Condition Reasoning

Cheng Yang · Jiaxuan Lu · Haiyuan Wan · Junchi Yu · Feiwei Qin

智能体与工具推理强化学习与后训练OpenReview
作者位次 #5/9ACL 2026Poster· Natural Language Processing

ACE -Router: Generalizing History-Aware Routing from MCP Tools to the Agent Web

Zhiyuan Yao · Zishan Xu · Yifu Guo · Zhiguang Han · Cheng Yang · Shuo Zhang · Ruiming Tang (Huawei Noah&#x27 · Xingshan Zeng · Weiwen Liu

摘要

With the rise of the Agent Web and Model Context Protocol (MCP), the agent ecosystem is evolving into an open collaborative network, exponentially increasing accessible tools. However, current architectures face severe scalability and generality bottlenecks. To address this, we propose ACE-Router...

智能体与工具评测与基准代码OpenReview 项目页
作者位次 #5/7ACL 2026Findings· Findings of ACL

R epo S hapley: Shapley-Enhanced Context Filtering for Repository-Level Code Completion

Yu Huo · Kun Zeng · Siyu Zhang · YuQuan Lu · Cheng Yang · Yifu Guo · Xiaoying Tang

摘要

Repository-level code completion benefits from retrieval-augmented generation (RAG). However, controlling cross-file evidence is difficult because chunk utility is often interaction-dependent: some snippets help only when paired with complementary context, while others harm decoding when they con...

代码检索与知识安全OpenReview 项目页
作者位次 #4/10ACL 2026Findings· Findings of ACL

From Experience to Skill: Multi-Agent Generative Engine Optimization via Reusable Strategy Learning

Beining Wu · Fuyou Mao · Jiong Lin · Cheng Yang · Jiaxuan Lu · Yifu Guo · Siyu Zhang · Yifan Wu · Ying Huang · Fu Li

摘要

Generative engines (GEs) are reshaping information access by replacing ranked links with citation-grounded answers, yet current Generative Engine Optimization (GEO) methods optimize each instance in isolation, unable to accumulate or transfer effective strategies across tasks and engines. We refr...

智能体与工具评测与基准推理OpenReview 项目页

常见合作者