PPaperPicks

Muzhi Dai

4 papers at tracked venues · 4 at CORE A* · active 2025–2025

DBLP profile ↗ORCID search ↗

Venues

ACM MM ×2AAAI ×1NeurIPS ×1

Papers

  1. CryoDomain: Sequence-free Protein Domain Identification from Low-resolution Cryo-EM Density Maps
    AAAI 2025 · Muzhi Dai
  2. From Captions to Rewards (CaReVL): Leveraging Large Language Model Experts for Enhanced Reward Modeling in Large Vision-Language Models
    ACM MM 2025 · Muzhi Dai
  3. S-GRPO: Early Exit via Reinforcement Learning in Reasoning Models
    NeurIPS 2025 · Muzhi Dai
  4. Secure Tug-of-War (SecTOW): Iterative Defense-Attack Training with Reinforcement Learning for Multimodal Model Security
    ACM MM 2025 · Muzhi Dai