A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Unknown
This paper proposes collaborative multi-agent debate for scalable oversight, improving error detection in LLM responses.
Unknown
This paper explores scalable oversight protocols, specifically evaluating whether weak LLMs can effectively judge strong LLMs, and expresses optimism about debate as a scalable oversight method.
Unknown
This paper introduces Slm-mux, a framework that orchestrates multiple small language models to improve reasoning performance, addressing the high failure rate of LLM-Debate with SLMs.
Junsol Kim, Shiyang Lai, Nino Scherrer, et al.
Reasoning models outperform instruction-tuned models not by longer chains of thought but by simulating multi-agent-like interactions that diversify and debate internal cognitive perspectives.
Unknown
Proposes a multi-agent debate framework where multiple language models iteratively critique and refine answers to improve factual accuracy and reasoning.
Unknown
Multiagent Finetuning improves LLMs by training specialized generation and critic agents on debate-generated data, enabling iterative self-improvement with diverse reasoning.