Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements Paper • 2608.17310 • Published 3 days ago • 96
SofT-GRPO: Surpassing Discrete-Token LLM Reinforcement Learning via Gumbel-Reparameterized Soft-Thinking Policy Optimization Paper • 2511.06411 • Published Nov 9, 2025 • 18
