5 papers
VEPO: Variable Entropy Policy Optimization for Low-Resource Language Foundation Models
Chonghan Liu, Yimin Du, Qi An +8
Large language models frequently exhibit suboptimal performance on low resource languages, primarily due to inefficient subword segmentation and systemic training data imbalances.…
ARC: Active and Reflection-driven Context Management for Long-Horizon Information Seeking Agents
Yilun Yao, Shan Huang, Elsie Dai +5
Large language models are increasingly deployed as research agents for deep search and long-horizon information seeking, yet their performance often degrades as interaction histori…
360-LLaMA-Factory: Plug & Play Sequence Parallelism for Long Post-Training
Haosheng Zou, Xiaowei Lv, Shousheng Jia +3
Adding sequence parallelism into LLaMA-Factory, we open-sourced 360-LLaMA-Factory at https://github.com/Qihoo360/360-LLaMA-Factory. 360-LLaMA-Factory has received wide recognition…
Light-IF: Endowing LLMs with Generalizable Reasoning via Preview and Self-Checking for Complex Instruction Following
Chenyang Wang, Liang Wen, Shousheng Jia +2
While advancements in the reasoning abilities of LLMs have significantly enhanced their performance in solving mathematical problems, coding tasks, and general puzzles, their effec…
Light-R1: Curriculum SFT, DPO and RL for Long COT from Scratch and Beyond
Liang Wen, Yunke Cai, Fenrui Xiao +11
This paper introduces Light-R1, an open-source suite for training long reasoning models using reproducible and cost-effective methodology. Given the proprietary nature of data used…