The papers behind how we build

All entries

5 entries

1. 2026PreprintCurated

   BM25 Wins at Scale: A Scaling Study of Retrieval-Augmented Generation Paradigms

   Wang et al.

   A controlled scaling study of retrieval-augmented generation paradigms, usually evaluated on different benchmarks at one corpus size. Corpus size is varied along 28 strictly nested tiers spanning roughly 450-fold while questions and a fixed bedrock of relevant and adversarial documents are held constant, under one reader model and one judging protocol. The result is a scale-dependent crossover rather than an unconditional winner. File-System Agent leads at the smallest shared tiers but costs 39 times more query tokens at the bedrock and becomes less effective as the search space grows. Around 10 million corpus tokens BM25 overtakes it and leads at every larger shared tier, with a margin approaching 20 points at full scale, and anchors the low-cost end of the Pareto frontier without any LLM-based construction. Dense retrieval remains efficient but less accurate; graph-based RAG encounters construction walls before deployment scale and its scalable variants stay below BM25 at shared tiers. The authors conclude that corpus growth increasingly favours global candidate ranking, with agentic reasoning working best after ranked discovery rather than in place of it.

   Retrieval & RAGEvaluation

   arXiv
2. 2023TACL 2024PublishedCurated

   Lost in the Middle: How Language Models Use Long Contexts

   Liu et al.

   Language models attend strongly to the beginning and end of a long context and weakly to the middle. The authors show accuracy degrading in a characteristic U-shape as the position of the relevant passage moves through the input, even in models explicitly built for long contexts.

   Retrieval & RAGEvaluation

   arXiv
3. 2021NeurIPS 2021PublishedCurated

   BEIR: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models

   Thakur et al.

   A collection of retrieval datasets spanning very different domains and task types, used to test whether a retriever that works on one corpus still works on another without retraining. The headline finding is that sparse lexical baselines remain hard to beat out of domain.

   Retrieval & RAGEvaluation

   arXiv
4. 2020NeurIPS 2020PublishedCurated

   Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

   Lewis et al.

   Pairs a pretrained generator with a neural retriever over an external corpus, so answers can draw on documents that were never in the training data and can be updated by changing the corpus rather than retraining the model.

   Retrieval & RAG

   arXiv
5. 2020EMNLP 2020PublishedCurated

   Dense Passage Retrieval for Open-Domain QA

   Karpukhin et al.

   Replaces term matching with a dual-encoder that embeds questions and passages into the same vector space. Relevance becomes a nearest-neighbour lookup, which lets retrieval find passages that share no vocabulary with the question.

   Retrieval & RAG

   arXiv

Browse by topic