Dong Wang

6 papers Meta Platforms Inc.

• Collaborative Reasoner: Self-Improving Social Agents with Synthetic Conversations
• Equilibrium Policy Generalization: A Reinforcement Learning Framework for Cross-Graph Zero-Shot Generalization in Pursuit-Evasion Games
• Hybrid Latent Reasoning via Reinforcement Learning
• LOMIA: Label-Only Membership Inference Attacks against Pre-trained Large Vision-Language Models
• Meta CLIP 2: A Worldwide Scaling Recipe
• NaturalReasoning: Reasoning in the Wild with 2.8M Challenging Questions