CoTAR: Chain-of-Thought Attribution Reasoning with Multi-level Granularity
This addresses the issue of unreliable attributions in LLM-based QA systems for users needing verifiable answers, representing an incremental improvement over existing attribution methods.
The paper tackles the problem of LLMs hallucinating information in QA tasks by introducing an attribution-oriented Chain-of-Thought reasoning method to improve attribution accuracy, demonstrating improved accuracy and correctness on two datasets using GPT-4 and showing that combining it with finetuning can enable smaller LLMs to outperform GPT-4 in some cases.
State-of-the-art performance in QA tasks is currently achieved by systems employing Large Language Models (LLMs), however these models tend to hallucinate information in their responses. One approach focuses on enhancing the generation process by incorporating attribution from the given input to the output. However, the challenge of identifying appropriate attributions and verifying their accuracy against a source is a complex task that requires significant improvements in assessing such systems. We introduce an attribution-oriented Chain-of-Thought reasoning method to enhance the accuracy of attributions. This approach focuses the reasoning process on generating an attribution-centric output. Evaluations on two context-enhanced question-answering datasets using GPT-4 demonstrate improved accuracy and correctness of attributions. In addition, the combination of our method with finetuning enhances the response and attribution accuracy of two smaller LLMs, showing their potential to outperform GPT-4 in some cases.