Rethinking with Retrieval: Faithful Large Language Model Inference
By Hangfeng He, Hongming Zhang, Dan Roth. 2022
217 citations recorded by the source corpus.
Despite the success of large language models (LLMs) in various natural language processing (NLP) tasks, the stored knowledge in these models may inevitably be incomplete, out-of-date, or incorrect. This motivates the need to utilize external knowledge to assist LLMs. Unfortunately, current methods for incorporating external knowledge often require additional training or fine-tuning, which can be costly and may not be feasible for LLMs. To address this issue, we propose a novel post-processing approach, rethinking with retrieval (RR), which retrieves relevant external knowledge based on the decomposed reasoning steps obtained from the chain-of-thought (CoT) prompting. This lightweight approach does not require additional training or fine-tuning and is not limited by the input length of LLMs. We evaluate the effectiveness of RR through extensive experiments with GPT-3 on three complex reasoning tasks: commonsense reasoning, temporal reasoning, and tabular reasoning. Our results show that RR can produce more faithful explanations and improve the performance of LLMs.
Subfields: Artificial Intelligence, Retrieval-Augmented Generation
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mohit Bansal
- Jianfeng Gao
- Yejin Choi
- Luke S. Zettlemoyer
- Percy Liang
- Graham Neubig
- Hannaneh Hajishirzi
- Noah A. Smith
In Retrieval-Augmented Generation
- Danqi Chen
- Jianfeng Gao
- Edouard Grave
- Weizhu Chen
- Daniel Khashabi
- Pengcheng He
- Tianyu Gao
- Jiatong Yu
Citations in the corpus
- Augmented Language Models: a Survey (2023, background): “Recent works (He et al., 2022; Trivedi et al., 2022) propose to combine a retriever with reasoning via chain-of-thoughts (CoT) prompting to augment a LM.”
- Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback (2023, background): “More recent work attempts to combine black-box LLMs with external knowledge, such as incorporating external knowledge into prompts (Madaan et al., 2022; Lazaridou et al., 2022), making GPT-3 more faithful (He et al., 2022), and combining web knowledge with GPT-3 (Nakano et al., 2021).”
- Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning (2023, methodology): “Therefore, the KL divergence can be calculated as where we use the shortest paths Z ∗ ⊆ Z between e q and e a in KGs as supervision signals (Zhang et al., 2022).”
- “According to . . . ”: Prompting Language Models Improves Quoting from Pre-Training Data (2023, methodology): “…documents before generation (Sun et al., 2023; Borgeaud et al., 2022; Mallen et al., 2023; Weller et al., 2022), applying new decoding approaches (He et al., 2022), post hoc tuning of LLMs (Menick et al., 2022; Lee et al., 2022b), and analyzing the model’s output training data (Han and Tsvetkov,…”
- A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions (2023, unknown): “…the substantial advancements chain-of-thought prompting has brought to LLMs in multi-step reasoning Wei et al. (2022), numerous studies (Yao et al., 2022; Trivedi et al., 2023; He et al., 2023) Beyond multi-step reasoning tasks, Jiang et al. (2023) shift their emphasis to long-form generation.”
- Large Language Models for Information Retrieval: A Survey (2023, unknown): “RAG [167] follows this strategy but applies the generative language modeling paradigm.”
- Enabling Large Language Models to Generate Text with Citations (2023, background): “Gao et al. (2023); He et al. (2022) propose to first generate text without accessing external documents and then retrieve relevant documents and revise the generation to be consistent.”
- ExpertQA: Expert-Curated Questions and Attributed Answers (2023, methodology): “Finally, post-hoc retrieval (Gao et al., 2023; He et al., 2022) involves retrieving attributions after answering a query using both the query and response for retrieval, and optionally revising the answer based on the attribution.”
- Benchmarking Large Language Models in Retrieval-Augmented Generation (2023, background): “…Guo et al. 2023), they still suffer severely from challenges including factual hallucination (Cao et al. 2020; Raunak, Menezes, and Junczys-Dowmunt 2021; Ji et al. 2023), knowledge out-dating (He, Zhang, and Roth 2022), and the lack of domain-specific expertise (Li et al. 2023c; Shen et al. 2023).”
- A Comprehensive Evaluation of Tool-Assisted Generation Strategies (2023, methodology): “Importantly, this implementation yields similar performance to that of existing approaches that combine CoT with retrieval from Wikipedia by He et al. (2022) 4) implemented methods that apply retrieval and refinement over generated CoT that are similar to C&F and achieve similar performance to…”
- Towards Reasoning in Large Language Models: A Survey (2022, background): “He et al. (2023) attempt to retrieve external knowledge in CoT to improve faithfulness of reasoning.”