# Blog

Occasional notes on research, ideas, and things I find worth writing down.

• On cross-model-family collaboration
• Mutual Reinforcement Learning: Sharing experience across heterogeneous LLM policies
• A note on copyright protection in the LLM era
• Both sides of LLM peer review at ICML 2026
• The Vision Wormhole: Latent-space communication for heterogeneous multi-agent systems
• Don't beg. Earn it. On a line from Eureka Seven
• The Trojan in the Vocabulary: How a single token can sabotage LLM composition