Backward Pass
@backwardpass.bsky.social
Ensembling LLMs with sample and voting is found to improve response accuracy asymptomatically.
arxiv.org/abs/2402.05120
arxiv.org/abs/2402.05120
More Agents Is All You Need
We find that, simply via a sampling-and-voting method, the performance of large language models (LLMs) scales with the number of agents instantiated. Also, this method is orthogonal to existing...
arxiv.org
February 9, 2024 at 2:51 PM
Ensembling LLMs with sample and voting is found to improve response accuracy asymptomatically.
arxiv.org/abs/2402.05120
arxiv.org/abs/2402.05120