Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312052651> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W4312052651 abstract "Recently, with the chain of thought (CoT) prompting, large language models (LLMs), e.g., GPT-3, have shown strong reasoning ability in several natural language processing tasks such as arithmetic, commonsense, and logical reasoning. However, LLMs with CoT require multi-step prompting and multi-token prediction, which is highly sensitive to individual mistakes and vulnerable to error accumulation. The above issues make the LLMs need the ability to verify the answers. In fact, after inferring conclusions in some thinking decision tasks, people often check them by re-verifying steps to avoid some mistakes. In this paper, we propose and prove that LLMs also have similar self-verification abilities. We take the conclusion obtained by CoT as one of the conditions for solving the original problem. By taking turns masking the original conditions and predicting their results, we calculate an explainable answer verification score based on whether the re-predicted conditions are correct. Experimental results demonstrate that the proposed method can improve the reasoning performance on various arithmetic, commonsense, and logical reasoning datasets. Our code is publicly available at: https://github.com/WENGSYX/Self-Verification." @default.
- W4312052651 created "2023-01-04" @default.
- W4312052651 creator A5024970317 @default.
- W4312052651 creator A5043842349 @default.
- W4312052651 creator A5045570568 @default.
- W4312052651 creator A5052553582 @default.
- W4312052651 creator A5076887655 @default.
- W4312052651 date "2022-12-19" @default.
- W4312052651 modified "2023-09-25" @default.
- W4312052651 title "Large Language Models are Better Reasoners with Self-Verification" @default.
- W4312052651 doi "https://doi.org/10.48550/arxiv.2212.09561" @default.
- W4312052651 hasPublicationYear "2022" @default.
- W4312052651 type Work @default.
- W4312052651 citedByCount "0" @default.
- W4312052651 crossrefType "posted-content" @default.
- W4312052651 hasAuthorship W4312052651A5024970317 @default.
- W4312052651 hasAuthorship W4312052651A5043842349 @default.
- W4312052651 hasAuthorship W4312052651A5045570568 @default.
- W4312052651 hasAuthorship W4312052651A5052553582 @default.
- W4312052651 hasAuthorship W4312052651A5076887655 @default.
- W4312052651 hasBestOaLocation W43120526511 @default.
- W4312052651 hasConcept C142362112 @default.
- W4312052651 hasConcept C153349607 @default.
- W4312052651 hasConcept C154945302 @default.
- W4312052651 hasConcept C177264268 @default.
- W4312052651 hasConcept C193221554 @default.
- W4312052651 hasConcept C199360897 @default.
- W4312052651 hasConcept C204321447 @default.
- W4312052651 hasConcept C2776760102 @default.
- W4312052651 hasConcept C2777402240 @default.
- W4312052651 hasConcept C38652104 @default.
- W4312052651 hasConcept C41008148 @default.
- W4312052651 hasConcept C43971567 @default.
- W4312052651 hasConcept C48145219 @default.
- W4312052651 hasConceptScore W4312052651C142362112 @default.
- W4312052651 hasConceptScore W4312052651C153349607 @default.
- W4312052651 hasConceptScore W4312052651C154945302 @default.
- W4312052651 hasConceptScore W4312052651C177264268 @default.
- W4312052651 hasConceptScore W4312052651C193221554 @default.
- W4312052651 hasConceptScore W4312052651C199360897 @default.
- W4312052651 hasConceptScore W4312052651C204321447 @default.
- W4312052651 hasConceptScore W4312052651C2776760102 @default.
- W4312052651 hasConceptScore W4312052651C2777402240 @default.
- W4312052651 hasConceptScore W4312052651C38652104 @default.
- W4312052651 hasConceptScore W4312052651C41008148 @default.
- W4312052651 hasConceptScore W4312052651C43971567 @default.
- W4312052651 hasConceptScore W4312052651C48145219 @default.
- W4312052651 hasLocation W43120526511 @default.
- W4312052651 hasOpenAccess W4312052651 @default.
- W4312052651 hasPrimaryLocation W43120526511 @default.
- W4312052651 hasRelatedWork W1985412924 @default.
- W4312052651 hasRelatedWork W2123996664 @default.
- W4312052651 hasRelatedWork W2375389409 @default.
- W4312052651 hasRelatedWork W2488051804 @default.
- W4312052651 hasRelatedWork W2611614995 @default.
- W4312052651 hasRelatedWork W2810280135 @default.
- W4312052651 hasRelatedWork W2950896474 @default.
- W4312052651 hasRelatedWork W2952018704 @default.
- W4312052651 hasRelatedWork W4288329983 @default.
- W4312052651 hasRelatedWork W4312052651 @default.
- W4312052651 isParatext "false" @default.
- W4312052651 isRetracted "false" @default.
- W4312052651 workType "article" @default.