AkihikoWatanabe / paper_notes

たまに追加される論文メモ
https://AkihikoWatanabe.github.io/paper_notes
15 stars 0 forks source link

Boosting Theory-of-Mind Performance in Large Language Models via Prompting, Moghaddam+, Johns Hopkins University, arXiv'23 #581

Open AkihikoWatanabe opened 1 year ago

AkihikoWatanabe commented 1 year ago

https://arxiv.org/pdf/2304.11490.pdf

AkihikoWatanabe commented 1 year ago

LLMはTheory-of-mind reasoningタスクが苦手なことが知られており、特にzero shotでは非常にパフォーマンスが低かった。ToMタスクとは、エージェントの信念、ゴール、メンタルstate、エージェントが何を知っているか等をトラッキングすることが求められるタスクのこと。このようなタスクはLLMが我々の日常生活を理解する上で重要。

image

↑のToM Questionのシナリオと問題 Scenario: "The morning of the high school dance Sarah placed her high heel shoes under her dress and then went shopping. That afternoon, her sister borrowed the shoes and later put them under Sarah's bed." Question: When Sarah gets ready, does she assume her shoes are under her dress?

しかし、Zero shot CoTのようなstep by step thinking, CoTを適切に行うことで、OpenAIの直近3つのモデルのAccuracyが80%を超えた。特に、GPT4は100%のAccuracyを達成。人間は87%だった。

この結果は、少なくとのこの論文でテストしたドメインではLLMのsocial reasoningのパフォーマンスをどのようにブーストするかを示しており、LLMのbehaviorは複雑でsensitiveであることを示唆している。