▲ ScienceQA 数据集的一个例子要回答上图所示的例子,我们首先要回忆关于力的定义:「A force is a push or a pull that ... The direction of a push is ... The direction of a pull is ... 」,然后形成一个多步的推理过程:「The baby’s hand applies a force to the cabinet door. → This force causes the door to open. → The direction of this force is toward the baby’s hand. 」,最终得到正确答案:「This force is a pull. 」。 在 ScienceQA 任务中,模型需要在预测答案的同时输出详细地解释。在本文中,作者利用大规模语言模型生成背景知识和解释,作为一种思维链(CoT)来模仿人类具有的多步推理能力。 实验表明,目前的多模态问答方法在 ScienceQA 任务不能取得很好的表现。相反,通过基于思维链的提示学习,GPT-3 模型能在 ScienceQA 数据集上取得 75.17% 的准确率,同时可以生成质量较高的解释:根据人类评估,其中 65.2% 的解释相关、正确且完整。思维链也可以帮助 UnifiedQA 模型在 ScienceQA 数据集上取得 3.99% 的提升。
论文标题:
Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering
[1] Pan Lu, Swaroop Mishra, Tony Xia, Liang Qiu, Kai-Wei Chang, Song-Chun Zhu, Oyvind Tafjord, Peter Clark, Ashwin Kalyan, et al. Learn to explain: multimodal reasoning via thought chains for science question answering. In Advances in neural information processing systems (NeurIPS), 2022. [2] Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Ed Chi, Quoc Le, and Denny Zhou. Chain of thought prompting elicits reasoning in large language models. arXiv preprint arXiv:2201.11903, 2022.[3] Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language models are few-shot learners. In Advances in neural information processing systems (NeurIPS), 2020.[4] Daniel Khashabi, Sewon Min, Tushar Khot, Ashish Sabharwal, Oyvind Tafjord, Peter Clark, and Hannaneh Hajishirzi. UnifiedQA: Crossing format boundaries with a single qa system. In Findings of the Association for Computational Linguistics (EMNLP), 2020.[5] Aniruddha Kembhavi, Minjoon Seo, Dustin Schwenk, Jonghyun Choi, Ali Farhadi, and Hannaneh Hajishirzi. Are you smarter than a sixth grader? textbook question answering for multimodal machine comprehension. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.