Researcher — paper.dinq.me

2026 顶会论文

作者位次 #4/6ICML 2026Poster· Deep Learning / Theory

Balancing Learning Rates Across Layers: Exact Two-Step Dynamics and Optimal Scaling in Linear Neural Networks

Tianyu Pang · Vignesh Kothapalli · Shenyang Deng · Haohui Wang · Dawei Zhou · Yaoqing Yang

理论与学习动力学OpenReview

作者位次 #3/6ACL 2026Workshop· ACL System Demonstrations

P hy V er: Physics-Grounded Material Claim Verification with Multi-Fidelity Physical Evidence

Jianpeng Chen · Wangzhi Zhan · Haohui Wang · Brian Mayer · Dongqi Fu · Dawei Zhou

摘要

Material claims in papers, patents, etc., often involve physical feasibility (e.g., stability under conditions, property consistency), not just textual feasibility. Yet most claim verifiers operate over language, therefore producing ungrounded judgments. On the other hand, directfirst-principles...

可解释性多模态多语言与语言学OpenReview 项目页

作者位次 #3/5KDD 2026Oral· Knowledge Discovery & Data Mining

DuetDA: Decomposed and Dynamic Data Attribution with Model-State Gating for Accelerated Scientific Endeavors

Jianpeng Chen · Wangzhi Zhan · Haohui Wang · Dongqi Fu · Dawei Zhou

检索与知识OpenReview

Haohui Wang

获取完整研究档案

联系方式

2026 顶会论文

Balancing Learning Rates Across Layers: Exact Two-Step Dynamics and Optimal Scaling in Linear Neural Networks

P hy V er: Physics-Grounded Material Claim Verification with Multi-Fidelity Physical Evidence

DuetDA: Decomposed and Dynamic Data Attribution with Model-State Gating for Accelerated Scientific Endeavors

常见合作者