Researcher — paper.dinq.me

2026 顶会论文

作者位次 #5/5ICLR 2026Oral

Taming Momentum: Rethinking Optimizer States Through Low-Rank Approximation

Zhengbo Wang · Jian Liang · Ran He · Zilei Wang · Tieniu Tan

效率与系统OpenReview

作者位次 #4/4ICLR 2026Poster· Social Aspects / Privacy-preserving Statistics and Machine Learning

Stop Tracking Me! Proactive Defense Against Attribute Inference Attack in LLMs

Dong Yan · Jian Liang · Ran He · Tieniu Tan

安全效率与系统社会与人本OpenReview

作者位次 #6/6ACL 2026Poster· Natural Language Processing

What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time

Dong Yan · Jian Liang · Yanbo Wang · Shuo Lu · Ran He · Tieniu Tan

摘要

Test-Time Reinforcement Learning (TTRL) enables Large Language Models (LLMs) to enhance reasoning capabilities on unlabeled test streams by deriving pseudo-rewards from majority voting consensus.However, existing TTRL methods rely exclusively on positive pseudo-labeling strategies.Such reliance b...

推理可解释性多模态OpenReview 项目页

Tieniu Tan

获取完整研究档案

联系方式

2026 顶会论文

Taming Momentum: Rethinking Optimizer States Through Low-Rank Approximation

Stop Tracking Me! Proactive Defense Against Attribute Inference Attack in LLMs

What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time

常见合作者