activity
20242026
collaborators

22 papers

cs.CL2026

SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs

Kejian Zhu, Zhuoran Jin, Shangqing Tu +5

Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) exhibit fundamentally different behaviors in enhancing multi-task reasoning for large language models (LLMs). Our preli…

cs.CV2026

Beyond Simply Environment Scaling: Designing Effective Environment Distributions for Multimodal Agent Learning

Kejian Zhu, Zhuoran Jin, Dongqi Huang +4

Recent works train agents by constructing large-scale multimodal environment pools. However, we find that simply increasing the number of multimodal environments does not always be…

cs.CL2026

The Physics of Multi-Turn Long-Horizon Planning: From Pre-training to Post-training via Single- and Multi-Teacher On-Policy Agentic Distillation

Tianyi Men, Zhuoran Jin, Kang Liu +1

Multi-turn long-horizon planning is critical for foundation model agents, yet how to fundamentally improve it remains unclear. Existing models are trained on uncontrollable and opa…

cs.CL2026

Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning

Tianyi Men, Zhuoran Jin, Pengfei Cao +3

Multimodal web agents can assist humans in operating repetitive GUI tasks, where effective task planning is essential for decomposing complex tasks into executable actions. While s…

cs.CL2026

Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It

Yupu Hao, Zhuoran Jin, Huanxuan Liao +2

Tool use enables large language models (LLMs) to perform complex tasks, and recent agentic reinforcement learning (RL) methods show promise for enhancing model capabilities. Howeve…

cs.CL2026

Look Light, Think Heavy: What Multimodal Chain-of-Thought Reasoning Can and Cannot Do

Zhuoran Jin, Kejian Zhu, Hongbang Yuan +5

Chain-of-Thought (CoT) has become a standard method for improving reasoning capabilities in large language models (LLMs) by eliciting step-by-step thinking, but its effectiveness i…