The authors create the first large-scale dataset and taxonomy of failure modes in multi-agent LLM systems to explain their limited performance gains.
The surprising effectiveness of ppo in cooperative multi-agent games.Advances in Neural Information Processing Systems, 35:24611–24624
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
other 1
citation-polarity summary
verdicts
UNVERDICTED 4representative citing papers
IBTS framework uses influence shaping to improve zero-shot human-machine teaming beyond partner diversity alone, with gains shown in Overcooked-AI simulations and a 30-subject human study.
Interactive IRL is cast as bi-level optimization with an inner loop learning expert rewards and an outer loop learning interaction policies, solved by the convergent BISIRL algorithm.
LeDRL integrates a lightweight LLM using structured prompts and a reflective evaluator with self-attention DRL to achieve over 17% higher task success rates and better convergence in edge computing task offloading.
citing papers explorer
-
Why Do Multi-Agent LLM Systems Fail?
The authors create the first large-scale dataset and taxonomy of failure modes in multi-agent LLM systems to explain their limited performance gains.
-
Beyond Partner Diversity: An Influence-Based Team Steering Framework for Zero-Shot Human-Machine Teaming
IBTS framework uses influence shaping to improve zero-shot human-machine teaming beyond partner diversity alone, with gains shown in Overcooked-AI simulations and a 30-subject human study.
-
Interactive Inverse Reinforcement Learning of Interaction Scenarios via Bi-level Optimization
Interactive IRL is cast as bi-level optimization with an inner loop learning expert rewards and an outer loop learning interaction policies, solved by the convergent BISIRL algorithm.
-
LLM-Enhanced Deep Reinforcement Learning for Task Offloading in Collaborative Edge Computing
LeDRL integrates a lightweight LLM using structured prompts and a reflective evaluator with self-attention DRL to achieve over 17% higher task success rates and better convergence in edge computing task offloading.