The authors create the first large-scale dataset and taxonomy of failure modes in multi-agent LLM systems to explain their limited performance gains.
Heterogeneous multi-agent reinforce- ment learning for zero-shot scalable collaboration
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.AI 2representative citing papers
Structured per-agent randomness via ranked masking in attention allows symmetric agents to break ties and coordinate, achieving perfect success on symmetric tasks where deterministic policies fail and enabling zero-shot transfer across team sizes.
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.
-
Randomness is sometimes necessary for coordination
Structured per-agent randomness via ranked masking in attention allows symmetric agents to break ties and coordinate, achieving perfect success on symmetric tasks where deterministic policies fail and enabling zero-shot transfer across team sizes.