Jiajun Chai
2 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
ZipRL introduces an adaptive context compression framework that significantly improves the performance and efficiency of LLMs in complex, multi-turn agent tasks by combining multi-granularity compression with Hindsight Response Replay.
This paper proposes two horizon-control strategies, Progressive OPD (POPD) and Truncated OPD (TOPD), demonstrating that full rollouts are often unnecessary for On-Policy Distillation, leading to significant improvements in training efficiency.
Papers
Are Full Rollouts Necessary for On-Policy Distillation?
Yaocheng Zhang, Jiajun Chai, Yuqian Fu, Songjun Tu +6 more
This paper proposes two horizon-control strategies, Progressive OPD (POPD) and Truncated OPD (TOPD), demonstrating that full rollouts are often unnecessary for On-Policy Distillation, leading to signi…