8 papers
Simple-OPD: Demystifying Warm-up for On-policy Distillation
Tao Liu, Taiqiang Wu, Mao Zheng +5
On-policy distillation (OPD) trains a student on its own rollouts with token-level supervision from teacher models, but its effectiveness can depend strongly on the warm-up stage b…
MuChator: Enabling Active Music Discovery via Conversational Music LLMs in Douyin Music
Jiahao Liang, Linzhi Huang, Xuannan Liu +6
Douyin Music, a large-scale platform with millions of daily users, adopts an immersive, feed-based discovery paradigm, where users passively explore music through continuous recomm…
GRP: Goal-Reversed Prompting for Zero-Shot Evaluation with LLMs
Mingyang Song, Mao Zheng, Xuan Luo
Pairwise LLM-as-a-judge evaluation asks the judge to identify the \emph{better} of two candidate answers. We study a one-line modification that asks for the \emph{worse} answer ins…
WorldReasonBench: Human-Aligned Stress Testing of Video Generators as Future World-State Predictors
Keming Wu, Yijing Cui, Wenhan Xue +11
Commercial video generation systems such as Seedance2.0 and Veo3.1 have rapidly improved, strengthening the view that video generators may be evolving into "world simulators." Yet…
The 3rd Place Solution of CCIR CUP 2025: A Framework for Retrieval-Augmented Generation in Multi-Turn Legal Conversation
Da Li, Zecheng Fang, Qiang Yan +2
Retrieval-Augmented Generation has made significant progress in the field of natural language processing. By combining the advantages of information retrieval and large language mo…
FastCuRL: Curriculum Reinforcement Learning with Stage-wise Context Scaling for Efficient Training R1-like Reasoning Models
Mingyang Song, Mao Zheng, Zheng Li +4
Improving training efficiency continues to be one of the primary challenges in large-scale Reinforcement Learning (RL). In this paper, we investigate how context length and the com…