2 papers
cs.AI2026
DMDIntel: Interpreting Large Language Models via Dynamic Mode Decomposition
Amogh Joshi, Animesh Mukherjee, Sergey Utyuzhnikov
In this work, we introduce DMDIntel which uses dynamic mode decomposition (DMD) to make the predictions made by LLMs in a classification task interpretable. It develops an input at…
cs.AI2026
Dynamic Governance of Multi-LLM Agent Systems for Collaborative Conversational Outcomes
Alexander Liss, Nicholas Desmond, Santiago Gil Gallego
When two LLM agents with structurally opposed objectives interact across multiple turns, the absence of a shared goal function produces not competition but collapse: the visitor ca…