14 papers
Decoupled Visual Processing: Efficient Multimodal Adaptation via Modality-Specific Transformer Substitution
Mingkuan Feng, Zhengqi Wen, Jianhua Tao
The paper introduces Decoupled Visual Processing, a method that replaces the upper decoder layers of a pretrained language model with a lightweight, trainable transformer block ded…
SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning
Jinyang Wu, Shuo Yang, Zhengxi Lu +8
The paper introduces SEED, a framework that extracts reusable natural-language skills from on-policy trajectories and distills them back into the policy to provide dense token-leve…
TACO: Tool-Augmented Credit Optimization for Agentic Tool Use
Mingkuan Feng, Jinyang Wu, Hao Gu +5
Agentic multimodal models perform diverse operations on an image via code and reason over the returned view, an effective paradigm for fine-grained visual question answering. Howev…
OPID: On-Policy Skill Distillation for Agentic Reinforcement Learning
Shuo Yang, Jinyang Wu, Zhengxi Lu +8
Outcome-based reinforcement learning provides a stable optimization backbone for language agents, but its sparse trajectory-level rewards provide little guidance on which intermedi…
Spark: Strategic Policy-Aware Exploration via Dynamic Branching for Long-Horizon Agentic Learning
Jinyang Wu, Shuo Yang, Changpeng Yang +4
Reinforcement learning has empowered large language models to act as intelligent agents, yet training them for long-horizon tasks remains challenging due to the scarcity of high-qu…
Maestro: Reinforcement Learning to Orchestrate Hierarchical Model-Skill Ensembles
Jinyang Wu, Guocheng Zhai, Ruihan Jin +7
The proliferation of large language models (LLMs) and modular skills has endowed autonomous agents with increasingly powerful capabilities. Existing frameworks typically rely on mo…