الباحثون

Bolin Ding

المنشورات 2

نسخة أولية وصول مفتوح

Playing social deduction games with reinforcement fine-tuned large language models

Lingzhe Zhang, Yunpeng Zhai, Tong Jia وآخرون · 2026

Reinforcement fine-tuning (RFT) is increasingly used in applications where large language models (LLMs) interact with humans and other agents. Here we use social deduction games to study how RFT changes LLMs' social behaviour. We let fine-tuned and base LLM agents play hidden-role games that require hidden-state infere …

نسخة أولية وصول مفتوح

Fine-Tuning on Self-Generated and Reward-Weighted Data: Learning Dynamics, Convergence Rates, and Benefits of Off-Policyness

Zhiwei Wang, Yanxi Chen, Yaliang Li وآخرون · 2026

We study the learning dynamics of fine-tuning a policy model on self-generated and reward-weighted data, with particular focus on a generalized version of REINFORCE -- referred to as RE(S) -- that updates the rollout distribution once every $S \ge 1$ gradient steps. Prior work in bandits and reinforcement learning has …

المؤلفون المشاركون