AI.info

Dequan Wang

Explore Dequan Wang on AI.info.

• CABLE: Extending the Reach of Memory Retrieval via Complementary Antecedent-Based Linking and Expansion
• Learning Native Continuation for Action Chunking Flow Policies
• AgencyBench: Benchmarking the Frontiers of Autonomous Agents in 1M-Token Real-World Contexts
• Saying the Unsaid: Revealing the Hidden Language of Multimodal Systems Through Telephone Games