CE-GPPO: Coordinating Entropy via Gradient-Preserving Clipping Policy Optimization in Reinforcement Learning (2026 · 7 citations)
The paper tackles unstable policy entropy when large language models are trained with RL for reasoning. PPO-style methods, including GRPO and DAPO, clip the importance sampling ratio, and this silently drops gradients from low-probability tokens. The authors split tokens by advantage sign and probability. They argue…
Entropy Ratio Clipping as a Soft Global Constraint for Stable Reinforcement Learning (2026 · 1 citation)
LLM post-training with reinforcement learning is usually off-policy, so updates use samples from older policies.
From Curated Data to Scalable Models: Continual Pre-training of Dense and MoE Large Language Models for Tibetan (2026 · 1 citation)
Mainstream LLMs are trained mostly on high-resource languages, so Tibetan remains poorly served despite its millions of speakers and cultural importance. The authors build an end-to-end pipeline to close this gap.
Good Reasoning Makes Good Demonstrations: Implicit Reasoning Quality Supervision via In-Context Reinforcement Learning (2026)
RLVR rewards every correct answer equally, so a model may also be reinforced for sloppy or lucky reasoning that happens to end in the right result.

Co-authors here

12 of 18 co-authors with a page, the most papers together first.

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

Papers of our library cite their works 9 times (of the 149,439 papers whose bibliographies we read); 5 citing papers with a page here →.

Updated 1 November 2026