Pengyi Li

Affiliations:

According to our database¹, Pengyi Li authored at least 3 papers between 2025 and 2026.

Collaborative distances:

Timeline

Book In proceedings Article PhD thesis Dataset Other

2026

Back to Basics: Revisiting Exploration in Reinforcement Learning for LLM Reasoning via Generative Probabilities.

[DOI]

CoRR, February, 2026

MaxInfo: A Training-Free Key-Frame Selection Method Using Maximum Volume for Enhanced Video Understanding.

[DOI]

Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2026

2025

Confidence Is All You Need: Few-Shot RL Fine-Tuning of Language Models.

[DOI]

CoRR, June, 2025