Papers with compact causal representation

    1 papers
    CAMO: An Agentic Framework for Automated Causal Discovery from Micro Behaviors to Macro Emergence in LLM Agent Simulations (2026.findings-acl)

    Copied to clipboard

    Challenge: LLM-empowered agent simulations generate rich, adaptive, and often nonlinear interaction patterns.
    Approach: They propose an automated Causal discovery framework for LLM agent simulations that converts mechanistic hypotheses into computable factors and learns a compact causal representation centered on an emergent target.
    Outcome: Experiments across four emergent settings demonstrate the promise of CAMO.

    What is GenGO?

    GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

    Information

    About
    Limitations