Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Improving Language Models via Plug-and-Play Retrieval Feedback

About

Large language models (LLMs) exhibit remarkable performance across various NLP tasks. However, they often generate incorrect or hallucinated information, which hinders their practical applicability in real-world scenarios. Human feedback has been shown to effectively enhance the factuality and quality of generated content, addressing some of these limitations. However, this approach is resource-intensive, involving manual input and supervision, which can be time-consuming and expensive. Moreover, it cannot be provided during inference, further limiting its practical utility in dynamic and interactive applications. In this paper, we introduce ReFeed, a novel pipeline designed to enhance LLMs by providing automatic retrieval feedback in a plug-and-play framework without the need for expensive fine-tuning. ReFeed first generates initial outputs, then utilizes a retrieval model to acquire relevant information from large document collections, and finally incorporates the retrieved information into the in-context demonstration for output refinement, thereby addressing the limitations of LLMs in a more efficient and cost-effective manner. Experiments on four knowledge-intensive benchmark datasets demonstrate our proposed ReFeed could improve over +6.0% under zero-shot setting and +2.5% under few-shot setting, compared to baselines without using retrieval feedback.

Wenhao Yu, Zhihan Zhang, Zhenwen Liang, Meng Jiang, Ashish Sabharwal• 2023

Related benchmarks

TaskDatasetResultRank
Multi-hop Question Answering2WikiMultihopQA
EM27.7
278
Multi-hop Question AnsweringHotpotQA
F1 Score38
221
Question AnsweringPopQA
Accuracy45.1
186
Question AnsweringTriviaQA--
85
Fact VerificationFEVER
Accuracy0.827
67
Question AnsweringNQ (Natural Questions)
EM39.6
55
Showing 6 of 6 rows

Other info

Follow for update