Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
To mitigate multi-agent degradation caused by misleading peers, the authors design BaRe-Mem, an online Bayesian reliability memory that dynamically tracks advisor credibility using the central agent's internal belief representations and interaction history. These reliability estimates dynamically modulate advisor influence and gate the decision between querying external advisors or relying on autonomous reasoning. Across nine benchmarks and six foundational models, BaRe-Mem consistently outperforms multi-agent debate and majority voting under heavy noise, maintaining performance strictly above solo reasoning baselines and improving worker routing on complex tasks like MuSiQue.
Multi-agent collaboration often performs worse than solo reasoning when peers hallucinate, but Bayesian tracking of advisor reliability rooted in internal belief states keeps collective performance strictly above autonomous baselines even under high misinformation.
In multi-agent systems, reliable consultation is challenging because advisor capabilities vary across tasks, and misleading information can make consultation worse than autonomous reasoning. We introduce BaRe-Mem, an online Bayesian reliability memory for multi-agent consultation. It estimates advisor reliability based on the central model's internal belief representations and updates these estimates from historical interactions. These estimates modulate the influence of advisor responses and guide the choice between consultation and autonomous reasoning. Across nine benchmarks and six central models, BaRe-Mem is more robust to misleading advisor information than debate and majority voting. On the more challenging tasks, it remains above autonomous reasoning across all tested misleading levels. Moreover, we extend the BaRe-Mem mechanism to worker allocation in agent teams. On the MuSiQue benchmark, BaRe-Mem improves task completion over routing by historical success counts and identifies capable workers earlier.