conftrace_

Binbin Zheng

3 papers · 2026–2026 · 1 conference · across top CS/AI conferences

Conferences

ACL (3)

Top co-authors

Jiaye Lin (2) Ke Zeng (2) Yangyi Fang (2) Xiaoliang Fu (2) Chaowen Hu (2) Zekai Shao (2) Lu Pan (2) Cong Qin (2) Xun Yang (1) Long Bai (1)

Keywords

large language model (3) policy optimization (2) reinforcement learning (2) importance sampling (1) probability mass (1) machine unlearning (1) probability gradient (1) local entropy maximization (1) prefix-aware localized unlearning (1) gradient utilization (1)

Papers

Maximizing Local Entropy Where It Matters: Prefix-Aware Localized LLM Unlearning ACL 2026 MASPO: Unifying Gradient Utilization, Probability Mass, and Signal Reliability for Robust and Sample-Efficient LLM Reasoning ACL 2026 From log 𝜋 to 𝜋: Taming Divergence in Soft Clipping via Bilateral Decoupled Decay of Probability Gradient Weight ACL 2026