Skill-Uni
Skill-Uni is a unified reinforcement learning framework designed to jointly optimize skill retrieval and utilization; it adopts a generative skill retrieval paradigm and achieves end-to-end optimization through two-stage training (warm-up instruction fine-tuning and GRPO reinforcement learning).
Timeline (last 90 days)
Skill-Uni is a unified reinforcement learning framework aimed at jointly optimizing skill retrieval and utilization.
Skill-Uni adopts a two-stage training strategy: the first stage is warm-up instruction fine-tuning, and the second stage uses the GRPO reinforcement learning algorithm.
Skill-Uni was tested on five skill-related benchmarks, improving on average by 66.59% over Qwen3-4B and by 34.82% on average over LLaMA3.1-8B.
Skill-Uni is compatible with a variety of skill retrieval engines, and users can choose different retrieval backends according to their needs.
All Facts (4)
Skill-Uni is a unified reinforcement learning framework aimed at jointly optimizing skill retrieval and utilization.
50%UnverifiedSkill-Uni adopts a two-stage training strategy: the first stage is warm-up instruction fine-tuning, and the second stage uses the GRPO reinforcement learning algorithm.
50%UnverifiedSkill-Uni was tested on five skill-related benchmarks, improving on average by 66.59% over Qwen3-4B and by 34.82% on average over LLaMA3.1-8B.
50%UnverifiedSkill-Uni is compatible with a variety of skill retrieval engines, and users can choose different retrieval backends according to their needs.
50%