RAG

• Retriever
• Knowledge Base
• Reader

Process RAG

• Retrieval-Augmented Generation for Large Language Models: A Survey Survey Paper

• PK-ICR: Persona-Knowledge Interactive Multi-Context Retrieval for Grounded Dialogue EMNLP 2023

• Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions ACL 2023

• Self-Knowledge Guided Retrieval Augmentation for Large Language Models EMNLP 2023

• DB-GPT: Empowering Database Interactions with Private Large Language Models

• Retrieval-Augmented Thought Process as Sequential Decision Making

Query Rewriter

• [Compressing Context to Enhance Inference Efficiency of Large Language Models]

Retriever

• [Active Retrieval Augmented Generation] EMNLP 2023 🔥🔥🔥 interesting and useful -> may can be used in dialogues

• I^3 Retriever: Incorporating Implicit Interaction in Pre-trained Language Models for Passage Retrieval

• Learning Retrieval Augmentation for Personalized Dialogue Generation EMNLP 2023

• [Divide and Conquer: Towards Better Embedding-based Retrieval for Recommender Systems from a Multi-task Perspective]

• [TRAVEL: Tag-Aware Conversational FAQ Retrieval via Reinforcement Learning] interesting and useful work

Multimodal

• [CAR: Consolidation, Augmentation and Regulation for Recipe Retrieval]

Multilingual

• Soft Prompt Decoding for Multilingual Dense Retrieval [blog]

Reader

• CHAIN-OF-NOTE: ENHANCING ROBUSTNESS IN RETRIEVAL-AUGMENTED LANGUAGE MODELS Tecent 🔥

• [Knowledgeable Preference Alignment for LLMs in Domain-specific Question Answering]

Database

• [Trends in Integration of Knowledge and Large Language Models: A Survey and Taxonomy of Methods, Benchmarks, and Applications]

Level of Granularity

• Dense X Retrieval: What Retrieval Granularity Should We Use? [code]

Structure of Data

Knowledge Graph

• Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph [code]

Different Application

• Tree of Clarifications: Answering Ambiguous Questions with Retrieval-Augmented Large Language Models EMNLP2023 using RAG to clarify ambiguous questions

• Divide and Conquer: Towards Better Embedding-based Retrieval for Recommender Systems from a Multi-task Perspective 🔥🔥🔥🔥🔥 this is very inspiring!

Datasets

• FRESHLLMS: REFRESHING LARGE LANGUAGE MODELS WITH SEARCH ENGINE AUGMENTATION

Some Problems

Lost in Middle

1. https://mp.weixin.qq.com/s/dahnrxGBaNIfEOzruxCbOw useful in practice, similiar with UniMS-RAG, the weights of different sources, the rerank processing for retrieved documents from different sources

Retrieve or Generate？

• Self-DC: When to retrieve and When to generate? Self Divide-and-Conquer for Compositional Unknown Questions 🔥🔥🔥🔥🔥 the benchmark and framework both are valuable and important.

• Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts for Open-Domain QA? 🔥 a good paper for analysis, not a method paper

• Retrieve Only When It Needs: Adaptive Retrieval Augmentation for Hallucination Mitigation in Large Language Models

• Self-Knowledge Guided Retrieval Augmentation for Large Language Models

• ACTIVERAG: Revealing the Treasures of Knowledge via Active Learning

Sub Functions

• BRANCH-SOLVE-MERGE IMPROVES LARGE LANGUAGE MODEL EVALUATION AND GENERATION combine_sub_qas in Self-DC

References

1. https://mp.weixin.qq.com/s/BlU3I6Ww3L8a0_Dxt0lztA

2. https://github.com/HKUST-AI-Lab/Awesome-LLM-with-RAG

3. https://mp.weixin.qq.com/s/Hy78rtJuJTehAJIC-HK2Rg (baichuan)

4. RAG in EMNLP2023 -> https://mp.weixin.qq.com/s/GUroCns3yB_voufgKCXSUg

5. latest RAG papers: https://recsys.substack.com/