A curated list of awesome resources for RAG (Retrieval Augmentation Generation) exploration.
-
RAG: Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks - An information retrieval augmented generation model that can be used for various knowledge-intensive NLP tasks. (Lewis, Patrick, et al. 2020)
-
CRAG - Corrective Retrieval Augmented Generation - Enhance the robustness of language model generation by evaluating and augmenting the relevance of retrieved documents through a an evaluator and large-scale web searches. (Shi-Qi Yan, Jia-Chen Gu, et al. 2024) (llamapack)
-
Dense x Retrieval: What Retrieval Grnaularity Should We Use? - Improve dense retrieval by using a more fine-grained retrieval granularity as known as Propositions. (Tong Chen, et al. 2023) (llamapack)
-
In-Context Learning for Extreme Multi-Label Classification - A retrieval-augmented generation model that can be used for extreme multi-label classification. (Karel, et al. 2021) (llamapack)
-
Self-Discover: Large Language Models Self-Compose Reasoning Structures - A retrieval-augmented generation model that can be used for self-composing reasoning structures. (Pei, et al. 2021) (llamapack)
-
SELF-RAG: LEARNING TO RETRIEVE, GENERATE, AND CRITIQUE THROUGH SELF-REFLECTION - A retrieval-augmented generation model that can be used for self-reflection. (Akari, et al. 2023) (llamapack)
-
Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding - A retrieval-augmented generation model that can be used for table understanding. (Zilong, et al. 2024) (llamapack)
-
RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval - An approach to enhance RAG by creating a summary tree from text chunks, providing deeper insights and overcoming the limitations of short, contiguous text retrieval. (Sarthi, Parth, et al. 2024)
-
HiQA: A Hierarchical Contextual Augmentation RAG for Massive Documents QA - An advanced multi-document question-answering framework that integrates cascading metadata and a multi-route retrieval mechanism, enhancing the accuracy of RAG pipeline. (Chen, Xinyue, et al. 2024)
-
ActiveRAG: Revealing the Treasures of Knowledge via Active Learning - Enhances RAG by active learning to deepen LLMs' understanding of external knowledge through innovative Knowledge Construction and Cognitive Nexus mechanisms. (Xu, Zhipeng, et al. 2024)
-
Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity - The paper proposes an adaptive question-answering framework that dynamically selects the most suitable strategy for retrieval-augmented large language models based on the complexity of the query, using a classifier trained on automatically collected labels. (Jeong, Soyeong, et al. 2024)
-
RAG vs Fine-tuning: Pipelines, Tradeoffs, and a Case Study on Agriculture - RAG vs Fine-tuning case study on agriculture domain datasets. (Gupta, Aman, et al. 2024)
-
RA-DIT: Retrieval-Augmented Dual Instruction Tuning (RA-DIT) - Improve the performance of retrieval-augmented generation models by fine-tuning the retrieval and generation components jointly. (Khattar, Dheeraj, et al. 2021)
-
InstructRetro: Instruction Tuning Post Retrieval-Augmented Pretraining - A Large Language Model pretrained with retrieval before instruction tuning (Wei Ping, et al. 2023)
- Retrieval-Augmented Generation for Large Language Models: A Survey - This survey paper examines Retrieval-Augmented Generation (RAG) in Large Language Models, discussing three paradigms—Naive, Advanced, and Modular RAG—and their key components. It evaluates RAG's impact on accuracy, hallucination reduction, and knowledge updates, also suggesting future research directions for enhancing RAG's effectiveness and scalability. (Yunfan Gao, et al. 2023)
Interested in contributing? Please read the contribution guidelines first.