Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering

Abstract

When answering a question, humans utilize the information available across different modalities to synthesize a consistent and complete chain of thought (CoT). This process is normally a black box in the case of deep learning models like large-scale language models. Recently, science question benchmarks have been used to diagnose the multi-hop reasoning ability and interpretability of an AI system. However, existing datasets fail to provide annotations for the answers, or are restricted to the textual-only modality, small scales, and limited domain diversity. To this end, we present Science Question Answering (ScienceQA), a new benchmark that consists of ~21k multimodal multiple choice questions with a diverse set of science topics and annotations of their answers with corresponding lectures and explanations. We further design language models to learn to generate lectures and explanations as the chain of thought (CoT) to mimic the multi-hop reasoning process when answering ScienceQA questions. ScienceQA demonstrates the utility of CoT in language models, as CoT improves the question answering performance by 1.20% in few-shot GPT-3 and 3.99% in fine-tuned UnifiedQA. We also explore the upper bound for models to leverage explanations by feeding those in the input; we observe that it improves the few-shot performance of GPT-3 by 18.96%. Our analysis further shows that language models, similar to humans, benefit from explanations to learn from fewer data and achieve the same performance with just 40% of the data. The data and code are available at https://scienceqa.github.io.

arXiv commentAccepted to NeurIPS 2022. 22 pages, 17 figures, 9 tables. Project: https://scienceqa.github.io

Cited by in corpus (12)

• Thrilled by Your Progress! Large Language Models (GPT-4) No Longer Struggle to Pass Assessments in Higher Education Programming Courses
• Can GPT-4 Support Analysis of Textual Data in Tasks Requiring Highly Specialized Domain Expertise?
• Can Generative Pre-trained Transformers (GPT) Pass Assessments in Higher Education Programming Courses?
• Language Model Crossover: Variation through Few-Shot Prompting
• In-Context Operator Learning with Data Prompts for Differential Equation Problems
• Complex QA and language models hybrid architectures, Survey
• MMInstruct: A High-Quality Multi-Modal Instruction Tuning Dataset with Extensive Diversity
• EvidenceMap: Learning Evidence Analysis to Unleash the Power of Small Language Models for Biomedical Question Answering
• Shotluck Holmes: A Family of Efficient Small-Scale Large Language Vision Models For Video Captioning and Summarization
• Sycophancy in Vision-Language Models: A Systematic Analysis and an Inference-Time Mitigation Framework
• MM-PoE: Multiple Choice Reasoning via. Process of Elimination using Multi-Modal Models
• CPR: Mitigating Large Language Model Hallucinations with Curative Prompt Refinement