PPaperPicks

Matthew Lyle Olson

9 papers at tracked venues · 7 at CORE A* · active 20242026

Venues

Frequent coauthors

Papers

  1. Bayesian Social Deduction with Graph-Informed Language Models
  2. Democratizing Diplomacy: A Harness for Evaluating Any Large Language Model on Full-Press Diplomacy
  3. LieCraft: A Multi-Agent Framework for Evaluating Deceptive Capabilities in Language Models
    AAAI 2026 · Matthew Lyle Olson
  4. Analyzing Hierarchical Structure in Vision Models with Sparse Autoencoders
    CVPR 2025 · Matthew Lyle Olson
  5. BLIP-3: A Family of Open Large Multimodal Models
  6. Debias Your Large Multi-Modal Model at Test-Time via Non-Contrastive Visual Attribute Steering
  7. Probing Semantic Routing in Large Mixture-of-Expert Models
    EMNLP 2025 · Matthew Lyle Olson
  8. Probing the Representational Power of Sparse Autoencoders in Vision Models
    ICCV 2025 · Matthew L. Olson
  9. Why do LLaVA Vision-Language Models Reply to Images in English?