Poster · Sydney

The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning

Name: The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning
Start: 2026-12-09T17:00:00+11:00
End: 2026-12-09T20:00:00+11:00
Location: Hall 1-4, ICC Sydney

When: Wed 9 Dec 2026, 17:00–20:00 AEDT
Where: Hall 1-4, ICC Sydney
Exhibition Centre, Level 1
Session: Sydney Poster Session 4
Authors: Jing Liang, Hongyao Tang, Yi Ma, Yancheng He, Weixun Wang, Xiaoyang Li, Wenbo Su, Jinyi Liu, YAN ZHENG, Jianye Hao, Bo Zheng
Organisations: Tianjin University, Mila, Université de Montréal, Shanxi University, Alibaba
QRS classification: Reinforcement learning, Language models

Open in schedule Official page