Retrieval-augmented generation for

By Kerfalla Ciss´e, A. Dantas

Subfields: 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 Retrieval-Augmented Generation

1. Jonathan Berant
2. Ohad Rubin
3. Jonathan Herzig

Citations in the corpus

• Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering (2020, neutral): “It proceeds in two steps, first retrieving support passages before processing them with a sequence to sequence model.”
• Autoregressive Search Engines: Generating Substrings as Document Identifiers (2022, neutral): “The former retrieves relevant context, usually in the form of short passages, which the latter then examines to produce answers (Chen et al., 2017; Lewis et al., 2020; Izacard and Grave, 2021).”
• Memory-assisted prompt editing to improve GPT-3 after deployment (2022, neutral): “cent success of injecting commonsense in the input (Lewis et al., 2020; Talmor et al., 2020), we propose a novel approach of injecting knowledge in the input via interactive feedback from an end-user.”
• KAT: A Knowledge Augmented Transformer for Vision-and-Language (2021, neutral): “Recent works (Izac-ard and Grave, 2020; Lewis et al., 2020) combine text retrieval models with language generative models which achieve state-of-the-art performance on knowledge-intensive natural language processing tasks.”
• Learning To Retrieve Prompts for In-Context Learning (2021, positive): “Work on retrieval-based methods has also spread more widely to other knowledge-intensive tasks (Lewis et al., 2020), e.”
• Entity-Based Knowledge Conflicts in Question Answering (2021, neutral): “We choose to evaluate a simple T5 reader model because it is the consistent component across highperforming retrieval-based QA models (Izacard and Grave, 2021; Lewis et al., 2020; Kim et al., 2020), and thus preserves the generality of our findings.”
• Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer (2022, unknown): “There have been several attempts to connect up reader and retriever training (Lee et al., 2019; Guu et al., 2020; Lewis et al., 2020; Sachan et al., 2021; Lee et al., 2021a; Izacard et al., 2022).”
• KG-FiD: Infusing Knowledge Graph in Fusion-in-Decoder for Open-Domain Question Answering (2021, unknown): “We also compare with methods (4) with additional largescale pre-training: REALM (Guu et al., 2020), RAG (Lewis et al., 2020) and Joint Top-K (Sachan et al., 2021).”
• Retrieval Augmented Visual Question Answering with Outside Knowledge (2022, unknown): “…to answer questions can be in pre-trained models (Roberts et al., 2020), knowledge-graphs (KGs) (Lin et al., 2019; Feng et al., 2020; Lv et al., 2020; Saffari et al., 2021) or document collections (Chen et al., 2017; Izacard and Grave, 2021; Guu et al., 2020; Lee et al., 2019; Lewis et al., 2020).”
• Knowledge-Augmented Methods for Natural Language Processing (2023, unknown): “…sequences, existing methods can be categorized into two methodologies: guiding generation with retrieved information (Ghazvininejad et al., 2018; Lewis et al., 2020; Wang et al., 2021); modeling background knowledge into text generation (Qin et al., 2019; Meng et al., 2020; Zeng et al., 2021).”
• Active Retrieval Augmented Generation (2023, unknown): “…based on the user’s input, and then generate a complete answer conditioning on the retrieved documents (Chen et al., 2017; Guu et al., 2020; Lewis et al., 2020; Izacard and Grave, 2021; Sachan et al., 2021; Lee et al., 2021; Jiang et al., 2022; Izacard et al., 2022; Nakano et al., 2021;…”
• RAGAs: Automated Evaluation of Retrieval Augmented Generation (2023, unknown): “The standard solution to these issues is to rely on Retrieval Augmented Generation (RAG) (Lee et al., 2019; Lewis et al., 2020; Guu et al., 2020).”
• Learning to repair: Repairing model output errors after deployment using a dynamic memory of feedback (2021, unknown): “…models that might generate an erroneous script which are correctable using an edit (feedback). ory to neural systems is not new, e.g., RAG (Lewis et al., 2020), REALM (Guu et al., 2020), ours is the first to utilize a memory of prior user feedback to improve future neural model performance.”
• Weakly-Supervised Visual-Retriever-Reader for Knowledge-based Question Answering (2021, unknown): “This observation is quite interesting as in the NLP domain, typically, a term-based retriever (e.g., TF-IDF and BM25) is a hard-to-beat baseline (Lee et al., 2019a; Lewis et al., 2020; Ma et al., 2021), suggesting an essential role of neural retrievers in the vision-&-language domain.”
• Boosted Dense Retriever (2021, unknown): “Identifying a small number of relevant documents from a large corpus to a given query, information retrieval is not only an important task in-and-of itself, but also plays a vital role in supporting a variety of knowledge-intensive NLP tasks (Lewis et al., 2020; Petroni et al., 2021), such as opendo”
• Phrase Retrieval Learns Passage Retrieval, Too (2021, unknown): “Recently, REALM (Guu et al., 2020) and RAG (retrieval-augmented generation) (Lewis et al., 2020) have been proposed as general-purpose pre-trained models with explicit access to world knowledge through the retriever.”
• Invoice Haystack: Benchmarking Document Retrieval and Visual Question Answering Under Strong Visual Homogeneity (2026, unknown): “Text-based RAG systems combine retrieval with generation: RAG [41] integrates DPR [31] and BART [40] for open-domain QA, REALM [26] incorporates retrieval into pre-training, Atlas [30] demonstrates strong few-shot performance through Fusion-in-Decoder [29], achieving 42.4% exact match on…”