Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
Paper
β’
2005.11401
β’
Published
β’
12
πβπ¬ RAG
Note The OG.
RAG vs Fine-tuning: Pipelines, Tradeoffs, and a Case Study on Agriculture
Paper β’ 2401.08406 β’ Published β’ 37Note This paper shows an example of generating a synthetic dataset.
-
BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models
Paper β’ 2104.08663 β’ Published β’ 3
Precise Zero-Shot Dense Retrieval without Relevance Labels
Paper β’ 2212.10496 β’ Published β’ 2Note This paper introduces HyDE method.
ReST meets ReAct: Self-Improvement for Multi-Step Reasoning LLM Agent
Paper β’ 2312.10003 β’ Published β’ 35Note This paper generates a synthetic dataset, with critique agents that filter out the generated examples based on relevance and groundedness.
-
Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agent
Paper β’ 2304.09542 β’ Published β’ 4 -
Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools
Paper β’ 2405.20362 β’ Published β’ 2