CoA w/o actions

Reported on 4 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing4 results

Question AnsweringonStrategyQA
EM· 2024-03-26
70.6
best: 79.2 (CoA)
Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models arXiv:2403.17359
Question AnsweringonWebQuestions
EM· 2024-03-26
64.7
best: 84.6 (PoG-GPT4 (Tan et al., 2024))
Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models arXiv:2403.17359
Question AnsweringonTruthfulQA
EM· 2024-03-26
63.3
best: 67.3 (CoA)
Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models arXiv:2403.17359
Question AnsweringonFEVER
EM· 2024-03-26
54.2
best: 68.9 (CoA)
Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models arXiv:2403.17359