Research Notes
Accessible explanations of the ideas behind my research.
2026
-
Multi-agent coordination
How do we scale AI agents without putting one agent in charge of everything?
Paper: A Responsibility-Centered Coordination Mechanism for Decentralized Multi-Agent Systems
Most multi-agent systems rely on a central orchestrator—but as the team grows, it can become a bottleneck.
Read the note → · Preprint · Paper link coming soon
-
Multi-agent safety
When Intelligent Agents Talk, Can Their Biases Become Stronger?
Paper: The Social Cost of Intelligence: Emergence, Propagation, and Amplification of Stereotypical Bias in Multi-Agent Systems
How stereotypical bias can emerge, spread, and intensify through interactions in multi-agent systems.
Read the note → · AACL-IJCNLP 2026 Findings
-
LLM alignment
Same Question, Different Profile: Why Do LLMs Treat Us Unequally?
Paper: TriAlign: Towards Universal Truth Consistency in Personalized LLM Alignment
Why personalized answers must not create different versions of reality—or quietly unequal access to truth.
Read the note → · EMNLP 2026