# Chronological index of RAG and its technical substrate

This is the complete chronological view of the primary-source registry as of
**2026-08-09**.  It complements the explanatory [chronology](chronology.md):
that chapter follows causal architectural transitions, while this index makes
all registered works—including parsing, chunking, sparse execution, ANN,
ranking, evaluation, privacy, security, multimodal retrieval, memory, and
serving—discoverable by first public date.

Dates are the earliest public date recorded in [`sources.json`](sources.json),
not necessarily the later proceedings year.  Status is explicit because an
influential preprint or industry report is not equivalent to a peer-reviewed
result.  Entries link to an original paper, official proceedings page, or
first-party program/report; the registry's title, date, venue, URL, status, and
topics are the canonical metadata.

## How to read the eras

1. **1971–2013: retrieval and indexing foundations.** Relevance feedback,
   probabilistic term specificity, the vector-space model, BM25, passage
   segmentation, relevance models, dynamic pruning, RRF, PQ/OPQ, and ANN
   establish the substrate later RAG systems inherit.
2. **2014–2019: differentiable memory and open-domain retrieve/read.** Memory
   Networks, DrQA, knowledge-grounded dialogue, graph QA, dense latent
   retrieval, kNN-LM, early BERT reranking, document expansion, DiskANN, and
   multi-hop benchmarks move external evidence into neural NLP.
3. **2020–2021: modern neural retrieval and retrieval-conditioned generation.**
   REALM, DPR, RAG, FiD, ColBERT, ANCE, KILT, RocketQA, learned sparse search,
   RETRO, BEIR, ScaNN, SPANN, and training/distillation work define the modern
   stack.
4. **2022–2023: generalization, instruction, long-form attribution, and
   retrieval-aware control.** Atlas, Contriever, E5/INSTRUCTOR/GTR, HyDE,
   FLARE, IRCoT, ALCE, Self-RAG, RAPTOR, and broader evaluation make retrieval
   more controllable and auditable.
5. **2024–2026: adaptive/agentic policies, graph and visual documents,
   reasoning-aware retrieval, memory, safety, and systems.** The frontier learns
   whether/when/how to retrieve and stop, while evaluation exposes citation,
   freshness, poisoning, privacy, multimodal, long-context, and production
   trade-offs.

Cross-paper scores are not comparable merely because methods appear in the same
year.  Corpus snapshots, qrels, retrieval depth, generators, prompts, models,
judges, and budgets differ.  Use this index to find evidence, then read the
mechanism and limitation analysis in the linked handbook chapter.

## Registry summary

The index contains **203 works**: 1 benchmark-program, 2 industry-report, 181 peer-reviewed, 19 preprint.

Most represented topic tags:

- `benchmark` — 26
- `multi-hop` — 12
- `dense-retrieval` — 12
- `generation` — 9
- `efficiency` — 9
- `reranking` — 9
- `ann` — 9
- `pretraining` — 8
- `rag` — 8
- `multimodal` — 8
- `long-context` — 8
- `graph` — 8
- `evaluation` — 8
- `memory` — 7
- `reasoning` — 7
- `reinforcement-learning` — 7
- `citations` — 7
- `embeddings` — 7
- `retrieval` — 6
- `distillation` — 6
- `learned-sparse` — 6
- `security` — 6
- `privacy` — 6
- `routing` — 5
- `visual-rag` — 5
- `chunking` — 5
- `late-interaction` — 4
- `compression` — 4
- `iterative-retrieval` — 4
- `robustness` — 4

## Retrieval and indexing foundations (1971–2013)

### 1971

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 1 | 1971 | [Relevance Feedback in Information Retrieval](https://doi.org/10.1137/1.9781611971817.3) | The SMART Retrieval System; peer-reviewed | `relevance-feedback`, `query-expansion` |
### 1972

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 2 | 1972 | [A Statistical Interpretation of Term Specificity and Its Application in Retrieval](https://doi.org/10.1108/eb026526) | Journal of Documentation; peer-reviewed | `tf-idf`, `sparse-retrieval` |
### 1975

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 3 | 1975 | [A Vector Space Model for Automatic Indexing](https://doi.org/10.1145/361219.361220) | Communications of the ACM; peer-reviewed | `vector-space`, `information-retrieval` |
### 1976

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 4 | 1976 | [Relevance Weighting of Search Terms](https://doi.org/10.1002/asi.4630270302) | JASIS; peer-reviewed | `probabilistic-ir`, `term-weighting` |
### 1994

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 5 | 1994-11 | [Okapi at TREC-3](https://pages.nist.gov/trec-browser/trec3/proceedings/) | TREC-3 / NIST SP 500-225; peer-reviewed | `bm25`, `sparse-retrieval` |
### 1997

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 6 | 1997 | [TextTiling: Segmenting Text into Multi-Paragraph Subtopic Passages](https://aclanthology.org/J97-1003/) | Computational Linguistics; peer-reviewed | `chunking`, `topic-segmentation` |
### 2001

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 7 | 2001 | [Relevance-Based Language Models](https://doi.org/10.1145/383952.383972) | SIGIR 2001; peer-reviewed | `rm3`, `pseudo-relevance-feedback`, `language-model-retrieval` |
### 2003

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 8 | 2003 | [Efficient Query Evaluation Using a Two-Level Retrieval Process](https://doi.org/10.1145/956863.956944) | CIKM 2003; peer-reviewed | `inverted-index`, `dynamic-pruning`, `wand` |
### 2009

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 9 | 2009 | [Reciprocal Rank Fusion Outperforms Condorcet and Individual Rank Learning Methods](https://plg.uwaterloo.ca/~gvcormac/cormacksigir09-rrf.pdf) | SIGIR 2009; peer-reviewed | `fusion`, `rrf` |
### 2011

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 10 | 2011 | [Faster Top-k Document Retrieval Using Block-Max Indexes](https://doi.org/10.1145/2009916.2010048) | SIGIR 2011; peer-reviewed | `inverted-index`, `dynamic-pruning` |
| 11 | 2011 | [Product Quantization for Nearest Neighbor Search](https://doi.org/10.1109/TPAMI.2010.57) | IEEE TPAMI; peer-reviewed | `ann`, `quantization`, `compression` |
### 2013

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 12 | 2013 | [Optimized Product Quantization for Approximate Nearest Neighbor Search](https://openaccess.thecvf.com/content_cvpr_2013/html/Ge_Optimized_Product_Quantization_2013_CVPR_paper.html) | CVPR 2013; peer-reviewed | `ann`, `quantization`, `rotation` |

## Differentiable memory and open-domain retrieve/read (2014–2019)

### 2014

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 13 | 2014-10-15 | [Memory Networks](https://arxiv.org/abs/1410.3916) | ICLR 2015; peer-reviewed | `memory`, `multi-hop` |
### 2015

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 14 | 2015-03-31 | [End-To-End Memory Networks](https://proceedings.neurips.cc/paper/2015/hash/8fb21ee7a2207526da55a679f0332de2-Abstract.html) | NeurIPS 2015; peer-reviewed | `memory`, `differentiable-retrieval` |
### 2016

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 15 | 2016 | [Efficient and Robust Approximate Nearest Neighbor Search Using Hierarchical Navigable Small World Graphs](https://doi.org/10.1109/TPAMI.2018.2889473) | IEEE TPAMI; peer-reviewed | `ann`, `hnsw`, `vector-index` |
### 2017

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 16 | 2017 | [Billion-scale Similarity Search with GPUs](https://arxiv.org/abs/1702.08734) | IEEE Big Data 2017; peer-reviewed | `ann`, `gpu`, `faiss` |
| 17 | 2017-03-31 | [Reading Wikipedia to Answer Open-Domain Questions](https://aclanthology.org/P17-1171/) | ACL 2017; peer-reviewed | `open-qa`, `tf-idf`, `reader` |
### 2018

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 18 | 2018 | [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://aclanthology.org/D18-1259/) | EMNLP 2018; peer-reviewed | `multi-hop`, `benchmark`, `supporting-facts` |
| 19 | 2018 | [Open Domain Question Answering Using Early Fusion of Knowledge Bases and Text](https://aclanthology.org/D18-1455/) | EMNLP 2018; peer-reviewed | `knowledge-graph`, `text`, `graph-neural-network` |
| 20 | 2018 | [Retrieve and Refine: Improved Sequence Generation Models for Dialogue](https://aclanthology.org/W18-5713/) | EMNLP SCAI 2018; peer-reviewed | `generation`, `dialogue` |
| 21 | 2018 | [Wizard of Wikipedia: Knowledge-Powered Conversational Agents](https://openreview.net/forum?id=r1l73iRqKm) | ICLR 2019; peer-reviewed | `dialogue`, `knowledge-grounding` |
### 2019

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 22 | 2019 | [Context-Aware Term Weighting for First Stage Passage Retrieval](https://doi.org/10.1145/3397271.3401204) | SIGIR 2020; peer-reviewed | `learned-sparse`, `term-weighting` |
| 23 | 2019 | [DiskANN: Fast Accurate Billion-point Nearest Neighbor Search on a Single Node](https://proceedings.neurips.cc/paper/2019/hash/09853c7fb1d3f8ee67a61b6bf4a7f8e6-Abstract.html) | NeurIPS 2019; peer-reviewed | `ann`, `disk`, `vector-index` |
| 24 | 2019 | [PullNet: Open Domain Question Answering with Iterative Retrieval on Knowledge Bases and Text](https://aclanthology.org/D19-1242/) | EMNLP-IJCNLP 2019; peer-reviewed | `knowledge-graph`, `iterative-retrieval` |
| 25 | 2019-01 | [Passage Re-ranking with BERT](https://arxiv.org/abs/1901.04085) | arXiv; preprint | `reranking`, `cross-encoder` |
| 26 | 2019-04 | [Document Expansion by Query Prediction](https://arxiv.org/abs/1904.08375) | arXiv; preprint | `document-expansion`, `sparse-retrieval` |
| 27 | 2019-06-01 | [Latent Retrieval for Weakly Supervised Open Domain Question Answering](https://aclanthology.org/P19-1612/) | ACL 2019; peer-reviewed | `dense-retrieval`, `latent-evidence`, `ict` |
| 28 | 2019-11-01 | [Generalization through Memorization: Nearest Neighbor Language Models](https://openreview.net/forum?id=HklBjCEKvH) | ICLR 2020; peer-reviewed | `language-modeling`, `non-parametric-memory` |

## Modern neural retrieval and RAG (2020–2021)

### 2020

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 29 | 2020 | [Accelerating Large-Scale Inference with Anisotropic Vector Quantization](https://proceedings.mlr.press/v119/guo20h.html) | ICML 2020; peer-reviewed | `ann`, `scann`, `quantization` |
| 30 | 2020 | [Adaptive Semiparametric Language Models](https://aclanthology.org/2021.tacl-1.22/) | TACL 2021; peer-reviewed | `memory`, `adaptive-fusion` |
| 31 | 2020 | [Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval](https://openreview.net/forum?id=zeFrfgyZln) | ICLR 2021; peer-reviewed | `dense-retrieval`, `ann-negatives`, `training` |
| 32 | 2020 | [ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT](https://doi.org/10.1145/3397271.3401075) | SIGIR 2020; peer-reviewed | `late-interaction`, `multi-vector` |
| 33 | 2020 | [Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Steps](https://aclanthology.org/2020.coling-main.580/) | COLING 2020; peer-reviewed | `multi-hop`, `benchmark`, `reasoning` |
| 34 | 2020 | [Document Ranking with a Pretrained Sequence-to-Sequence Model](https://aclanthology.org/2020.findings-emnlp.63/) | Findings EMNLP 2020; peer-reviewed | `reranking`, `t5`, `generative-ranking` |
| 35 | 2020 | [DocVQA: A Dataset for VQA on Document Images](https://openaccess.thecvf.com/content/WACV2021/html/Mathew_DocVQA_A_Dataset_for_VQA_on_Document_Images_WACV_2021_paper.html) | WACV 2021; peer-reviewed | `document-ai`, `visual-qa`, `benchmark` |
| 36 | 2020 | [KILT: a Benchmark for Knowledge Intensive Language Tasks](https://aclanthology.org/2021.naacl-main.200/) | NAACL 2021; peer-reviewed | `benchmark`, `provenance`, `wikipedia` |
| 37 | 2020 | [RocketQA: An Optimized Training Approach to Dense Passage Retrieval for Open-Domain Question Answering](https://aclanthology.org/2021.naacl-main.466/) | NAACL 2021; peer-reviewed | `dense-retrieval`, `denoising`, `cross-batch-negatives` |
| 38 | 2020-02-10 | [REALM: Retrieval-Augmented Language Model Pre-Training](https://proceedings.mlr.press/v119/guu20a.html) | ICML 2020; peer-reviewed | `pretraining`, `latent-retrieval`, `index-refresh` |
| 39 | 2020-04-10 | [Dense Passage Retrieval for Open-Domain Question Answering](https://aclanthology.org/2020.emnlp-main.550/) | EMNLP 2020; peer-reviewed | `dense-retrieval`, `contrastive-learning`, `faiss` |
| 40 | 2020-05-22 | [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://proceedings.neurips.cc/paper/2020/hash/6b493230205f780e1bc26945df7481e5-Abstract.html) | NeurIPS 2020; peer-reviewed | `rag`, `latent-documents`, `seq2seq` |
| 41 | 2020-06-26 | [Pre-training via Paraphrasing](https://proceedings.neurips.cc/paper/2020/hash/d6f1dd034aabde7657e6680444ceff62-Abstract.html) | NeurIPS 2020; peer-reviewed | `pretraining`, `multilingual`, `retrieval` |
| 42 | 2020-07-02 | [Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering](https://aclanthology.org/2021.eacl-main.74/) | EACL 2021; peer-reviewed | `fusion-in-decoder`, `multi-passage` |
| 43 | 2020-12-08 | [Distilling Knowledge from Reader to Retriever for Question Answering](https://openreview.net/forum?id=NTEz-6wysdb) | ICLR 2021; peer-reviewed | `distillation`, `reader-to-retriever` |
### 2021

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 44 | 2021 | [BEIR: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models](https://datasets-benchmarks-proceedings.neurips.cc/paper/2021/hash/65b9eea6e1cc6bb9f0cd2a47751a186f-Abstract-round2.html) | NeurIPS 2021 Datasets and Benchmarks; peer-reviewed | `retrieval`, `benchmark`, `zero-shot` |
| 45 | 2021 | [COIL: Revisit Exact Lexical Match in Information Retrieval with Contextualized Inverted List](https://aclanthology.org/2021.naacl-main.241/) | NAACL 2021; peer-reviewed | `learned-sparse`, `multi-vector`, `lexical` |
| 46 | 2021 | [Condenser: a Pre-training Architecture for Dense Retrieval](https://aclanthology.org/2021.emnlp-main.75/) | EMNLP 2021; peer-reviewed | `dense-retrieval`, `pretraining` |
| 47 | 2021 | [GPL: Generative Pseudo Labeling for Unsupervised Domain Adaptation of Dense Retrieval](https://aclanthology.org/2022.naacl-main.168/) | NAACL 2022; peer-reviewed | `domain-adaptation`, `synthetic-queries`, `distillation` |
| 48 | 2021 | [Large Dual Encoders Are Generalizable Retrievers](https://aclanthology.org/2022.emnlp-main.669/) | EMNLP 2022; peer-reviewed | `dense-retrieval`, `scaling`, `zero-shot` |
| 49 | 2021 | [MuSiQue: Multihop Questions via Single-hop Question Composition](https://aclanthology.org/2022.tacl-1.31/) | TACL 2022; peer-reviewed | `multi-hop`, `benchmark`, `composition` |
| 50 | 2021 | [SPANN: Highly-efficient Billion-scale Approximate Nearest Neighborhood Search](https://proceedings.neurips.cc/paper_files/paper/2021/hash/299dc35e747eb77177d9cea10a802da2-Abstract.html) | NeurIPS 2021; peer-reviewed | `ann`, `disk`, `inverted-file` |
| 51 | 2021 | [SPLADE: Sparse Lexical and Expansion Model for First Stage Ranking](https://doi.org/10.1145/3404835.3463098) | SIGIR 2021; peer-reviewed | `learned-sparse`, `expansion` |
| 52 | 2021 | [Unsupervised Corpus Aware Language Model Pre-training for Dense Passage Retrieval](https://aclanthology.org/2022.acl-long.203/) | ACL 2022; peer-reviewed | `dense-retrieval`, `pretraining`, `contrastive` |
| 53 | 2021-04 | [Efficiently Teaching an Effective Dense Retriever with Balanced Topic Aware Sampling](https://doi.org/10.1145/3404835.3462891) | SIGIR 2021; peer-reviewed | `dense-retrieval`, `distillation`, `sampling` |
| 54 | 2021-04 | [Learning Passage Impacts for Inverted Indexes](https://doi.org/10.1145/3459637.3482273) | CIKM 2021; peer-reviewed | `learned-sparse`, `impact-index` |
| 55 | 2021-06-09 | [End-to-End Training of Multi-Document Reader and Retriever for Open-Domain Question Answering](https://proceedings.neurips.cc/paper/2021/hash/da3fde159d754a2555eaa198d2d105b2-Abstract.html) | NeurIPS 2021; peer-reviewed | `joint-training`, `multi-document` |
| 56 | 2021-09 | [SPLADE v2: Sparse Lexical and Expansion Model for Information Retrieval](https://arxiv.org/abs/2109.10086) | arXiv; preprint | `learned-sparse`, `expansion` |
| 57 | 2021-12-03 | [ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction](https://aclanthology.org/2022.naacl-main.272/) | NAACL 2022; peer-reviewed | `late-interaction`, `compression` |
| 58 | 2021-12-08 | [Improving Language Models by Retrieving from Trillions of Tokens](https://proceedings.mlr.press/v162/borgeaud22a.html) | ICML 2022; peer-reviewed | `pretraining`, `chunk-retrieval`, `scaling` |
| 59 | 2021-12-16 | [Unsupervised Dense Information Retrieval with Contrastive Learning](https://openreview.net/forum?id=jKN1pXi7b0) | TMLR 2022; peer-reviewed | `dense-retrieval`, `unsupervised` |

## Generalization, attribution, and retrieval control (2022–2023)

### 2022

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 60 | 2022 | [ASQA: Factoid Questions Meet Long-Form Answers](https://aclanthology.org/2022.emnlp-main.566/) | EMNLP 2022; peer-reviewed | `long-form`, `ambiguous-qa`, `citations` |
| 61 | 2022 | [ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning](https://aclanthology.org/2022.findings-acl.177/) | Findings ACL 2022; peer-reviewed | `charts`, `visual-qa`, `reasoning` |
| 62 | 2022 | [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://proceedings.neurips.cc/paper/2022/hash/67d57c32e20fd0a7a302cb81d36e40d5-Abstract-Conference.html) | NeurIPS 2022; peer-reviewed | `attention`, `efficiency`, `serving` |
| 63 | 2022 | [Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions](https://aclanthology.org/2023.acl-long.557/) | ACL 2023; peer-reviewed | `multi-hop`, `iterative-retrieval`, `reasoning` |
| 64 | 2022 | [KG-FiD: Infusing Knowledge Graph in Fusion-in-Decoder for Open-Domain Question Answering](https://aclanthology.org/2022.acl-long.340/) | ACL 2022; peer-reviewed | `knowledge-graph`, `fid`, `multi-hop` |
| 65 | 2022 | [LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking](https://doi.org/10.1145/3503161.3548112) | ACM Multimedia 2022; peer-reviewed | `document-ai`, `layout`, `multimodal` |
| 66 | 2022 | [Matryoshka Representation Learning](https://proceedings.neurips.cc/paper_files/paper/2022/hash/c32319f4868da7613d78af9993100e42-Abstract-Conference.html) | NeurIPS 2022; peer-reviewed | `embeddings`, `adaptive-dimension`, `efficiency` |
| 67 | 2022 | [Measuring and Narrowing the Compositionality Gap in Language Models](https://openreview.net/forum?id=40yPtmzndN) | ICLR 2023; peer-reviewed | `self-ask`, `decomposition`, `search` |
| 68 | 2022 | [MTEB: Massive Text Embedding Benchmark](https://aclanthology.org/2023.eacl-main.148/) | EACL 2023; peer-reviewed | `embeddings`, `benchmark` |
| 69 | 2022 | [One Embedder, Any Task: Instruction-Finetuned Text Embeddings](https://aclanthology.org/2023.findings-acl.71/) | Findings ACL 2023; peer-reviewed | `embeddings`, `instructions` |
| 70 | 2022 | [RARR: Researching and Revising What Language Models Say, Using Language Models](https://aclanthology.org/2023.acl-long.910/) | ACL 2023; peer-reviewed | `verification`, `revision`, `attribution` |
| 71 | 2022 | [RetroMAE: Pre-Training Retrieval-oriented Language Models Via Masked Auto-Encoder](https://aclanthology.org/2022.emnlp-main.35/) | EMNLP 2022; peer-reviewed | `dense-retrieval`, `pretraining`, `masked-autoencoder` |
| 72 | 2022 | [SimLM: Pre-training with Representation Bottleneck for Dense Passage Retrieval](https://aclanthology.org/2023.acl-long.125/) | ACL 2023; peer-reviewed | `dense-retrieval`, `pretraining`, `bottleneck` |
| 73 | 2022 | [When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories](https://aclanthology.org/2023.acl-long.546/) | ACL 2023; peer-reviewed | `long-tail`, `parametric-memory`, `benchmark` |
| 74 | 2022-05 | [PLAID: An Efficient Engine for Late Interaction Retrieval](https://arxiv.org/abs/2205.09707) | arXiv; preprint | `late-interaction`, `indexing`, `efficiency` |
| 75 | 2022-05 | [QAMPARI: An Open-domain Question Answering Benchmark for Questions with Many Answers from Multiple Paragraphs](https://arxiv.org/abs/2205.12665) | arXiv; preprint | `list-qa`, `long-form`, `benchmark` |
| 76 | 2022-05 | [SPLADE++: Ensemble Distillation for High Performance Sparse Information Retrieval](https://arxiv.org/abs/2205.04733) | arXiv; preprint | `learned-sparse`, `distillation`, `hard-negatives` |
| 77 | 2022-08-05 | [Atlas: Few-shot Learning with Retrieval Augmented Language Models](https://jmlr.org/papers/v24/23-0037.html) | JMLR 2023; peer-reviewed | `pretraining`, `few-shot`, `distillation` |
| 78 | 2022-10-06 | [MuRAG: Multimodal Retrieval-Augmented Generator for Open Question Answering over Images and Text](https://aclanthology.org/2022.emnlp-main.375/) | EMNLP 2022; peer-reviewed | `multimodal`, `images`, `qa` |
| 79 | 2022-10-06 | [ReAct: Synergizing Reasoning and Acting in Language Models](https://openreview.net/forum?id=WE_vluYUL-X) | ICLR 2023; peer-reviewed | `agents`, `tool-use`, `reasoning` |
| 80 | 2022-11-22 | [Retrieval-Augmented Multimodal Language Modeling](https://arxiv.org/abs/2211.12561) | arXiv; preprint | `multimodal`, `generation` |
| 81 | 2022-12 | [Text Embeddings by Weakly-Supervised Contrastive Pre-training](https://arxiv.org/abs/2212.03533) | arXiv; preprint | `embeddings`, `weak-supervision`, `contrastive` |
| 82 | 2022-12-20 | [Precise Zero-Shot Dense Retrieval without Relevance Labels](https://aclanthology.org/2023.acl-long.99/) | ACL 2023; peer-reviewed | `query-transformation`, `hypothetical-document` |
### 2023

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 83 | 2023 | [AdANNS: A Framework for Adaptive Semantic Search](https://proceedings.neurips.cc/paper_files/paper/2023/hash/f062da1973ac9ac61fc6d44dd7fa309f-Abstract-Conference.html) | NeurIPS 2023; peer-reviewed | `ann`, `adaptive-representation`, `efficiency` |
| 84 | 2023 | [ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systems](https://aclanthology.org/2024.naacl-long.20/) | NAACL 2024; peer-reviewed | `evaluation`, `prediction-powered-inference` |
| 85 | 2023 | [CITADEL: Conditional Token Interaction via Dynamic Lexical Routing for Efficient and Effective Multi-Vector Retrieval](https://aclanthology.org/2023.acl-long.663/) | ACL 2023; peer-reviewed | `multi-vector`, `lexical-routing`, `efficiency` |
| 86 | 2023 | [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://doi.org/10.1145/3600006.3613165) | SOSP 2023; peer-reviewed | `serving`, `paged-attention`, `systems` |
| 87 | 2023 | [Enabling Large Language Models to Generate Text with Citations](https://aclanthology.org/2023.emnlp-main.398/) | EMNLP 2023; peer-reviewed | `citations`, `attributed-generation`, `benchmark` |
| 88 | 2023 | [FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation](https://aclanthology.org/2024.findings-acl.813/) | Findings ACL 2024; peer-reviewed | `freshness`, `search`, `dynamic-qa` |
| 89 | 2023 | [How to Train Your DRAGON: Diverse Augmentation Towards Generalizable Dense Retrieval](https://aclanthology.org/2023.findings-emnlp.423/) | Findings EMNLP 2023; peer-reviewed | `dense-retrieval`, `augmentation`, `generalization` |
| 90 | 2023 | [Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agents](https://aclanthology.org/2023.emnlp-main.923/) | EMNLP 2023; peer-reviewed | `reranking`, `llm`, `listwise` |
| 91 | 2023 | [LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models](https://aclanthology.org/2023.emnlp-main.825/) | EMNLP 2023; peer-reviewed | `prompt-compression`, `efficiency` |
| 92 | 2023 | [LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression](https://aclanthology.org/2024.acl-long.91/) | ACL 2024; peer-reviewed | `prompt-compression`, `long-context`, `position-bias` |
| 93 | 2023 | [RAGAS: Automated Evaluation of Retrieval Augmented Generation](https://aclanthology.org/2024.eacl-demo.16/) | EACL 2024 Demo; peer-reviewed | `evaluation`, `faithfulness` |
| 94 | 2023 | [RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models](https://aclanthology.org/2024.acl-long.585/) | ACL 2024; peer-reviewed | `hallucination`, `span-labels` |
| 95 | 2023 | [RankT5: Fine-Tuning T5 for Text Ranking with Ranking Losses](https://www.microsoft.com/en-us/research/publication/rankt5-fine-tuning-t5-for-text-ranking-with-ranking-losses/) | SIGIR 2023; peer-reviewed | `reranking`, `listwise`, `ranking-loss` |
| 96 | 2023 | [Worst-case Performance of Popular Approximate Nearest Neighbor Search Implementations: Guarantees and Limitations](https://proceedings.neurips.cc/paper_files/paper/2023/hash/d0ac28b79816b51124fcc804b2496a36-Abstract-Conference.html) | NeurIPS 2023; peer-reviewed | `ann`, `robustness`, `theory` |
| 97 | 2023-03 | [RepoCoder: Repository-Level Code Completion Through Iterative Retrieval and Generation](https://aclanthology.org/2023.emnlp-main.151/) | EMNLP 2023; peer-reviewed | `code`, `iterative-retrieval`, `generation` |
| 98 | 2023-03-14 | [Query2doc: Query Expansion with Large Language Models](https://arxiv.org/abs/2303.07678) | arXiv; preprint | `query-expansion`, `pseudo-document` |
| 99 | 2023-04 | [XTR: Rethinking the Role of Token Retrieval in Multi-Vector Retrieval](https://arxiv.org/abs/2304.01982) | arXiv; preprint | `multi-vector`, `token-retrieval`, `efficiency` |
| 100 | 2023-05-11 | [Active Retrieval Augmented Generation](https://aclanthology.org/2023.emnlp-main.495/) | EMNLP 2023; peer-reviewed | `active-retrieval`, `uncertainty` |
| 101 | 2023-05-23 | [Query Rewriting for Retrieval-Augmented Large Language Models](https://aclanthology.org/2023.emnlp-main.322/) | EMNLP 2023; peer-reviewed | `query-rewriting`, `reinforcement-learning` |
| 102 | 2023-05-24 | [Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergy](https://arxiv.org/abs/2305.15294) | arXiv; preprint | `iterative-retrieval`, `generation` |
| 103 | 2023-07-06 | [Lost in the Middle: How Language Models Use Long Contexts](https://aclanthology.org/2024.tacl-1.9/) | TACL 2024; peer-reviewed | `long-context`, `position`, `distraction` |
| 104 | 2023-08 | [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://arxiv.org/abs/2308.14508) | arXiv; preprint | `long-context`, `benchmark`, `bilingual` |
| 105 | 2023-08-25 | [Nougat: Neural Optical Understanding for Academic Documents](https://openreview.net/forum?id=fUtxNAKpdV) | ICLR 2024; peer-reviewed | `ocr`, `scientific-documents`, `parsing` |
| 106 | 2023-09-04 | [Benchmarking Large Language Models in Retrieval-Augmented Generation](https://ojs.aaai.org/index.php/AAAI/article/view/29728) | AAAI 2024; peer-reviewed | `evaluation`, `noise`, `counterfactual` |
| 107 | 2023-10 | [RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentation](https://openreview.net/forum?id=mlJLVigNHp) | ICLR 2024; peer-reviewed | `compression`, `selective-augmentation` |
| 108 | 2023-10-17 | [Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection](https://openreview.net/forum?id=hSyW5go0v8) | ICLR 2024; peer-reviewed | `adaptive-retrieval`, `reflection`, `citations` |
| 109 | 2023-12 | [Dense X Retrieval: What Retrieval Granularity Should We Use?](https://aclanthology.org/2024.emnlp-main.845/) | EMNLP 2024; peer-reviewed | `propositions`, `chunking`, `granularity` |

## Adaptive, multimodal, secure, and production RAG (2024–2026)

### 2024

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 110 | 2024 | [AgentPoison: Red-Teaming LLM Agents via Poisoning Memory or Knowledge Bases](https://proceedings.neurips.cc/paper_files/paper/2024/hash/eb113910e9c3f6242541c1652e30dfd6-Abstract-Conference.html) | NeurIPS 2024; peer-reviewed | `security`, `poisoning`, `backdoor` |
| 111 | 2024 | [C-RAG: Certified Generation Risks for Retrieval-Augmented Language Models](https://proceedings.mlr.press/v235/kang24a.html) | ICML 2024; peer-reviewed | `conformal-risk`, `certification` |
| 112 | 2024 | [CRAG: A Comprehensive RAG Benchmark](https://proceedings.neurips.cc/paper_files/paper/2024/hash/1435d2d0fca85a84d83ddcb754f58c29-Abstract-Datasets_and_Benchmarks_Track.html) | NeurIPS 2024 Datasets and Benchmarks; peer-reviewed | `benchmark`, `freshness`, `long-tail` |
| 113 | 2024 | [Docling Technical Report](https://arxiv.org/abs/2408.09869) | arXiv; preprint | `document-parsing`, `layout`, `tables` |
| 114 | 2024 | [Don't Forget Private Retrieval: Distributed Private Similarity Search for Large Language Models](https://aclanthology.org/2024.privatenlp-1.2/) | Privacy in NLP 2024; peer-reviewed | `privacy`, `mpc`, `similarity-search` |
| 115 | 2024 | [Generative Representational Instruction Tuning](https://proceedings.iclr.cc/paper_files/paper/2025/hash/70cfb8e8c9e142e87e33f329be4ddf86-Abstract-Conference.html) | ICLR 2025; peer-reviewed | `embeddings`, `generation`, `efficiency` |
| 116 | 2024 | [HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models](https://proceedings.neurips.cc/paper_files/paper/2024/hash/6ddc81d76dc3e20c1cdbda4a040d11ae-Abstract-Conference.html) | NeurIPS 2024; peer-reviewed | `graph`, `pagerank`, `multi-hop` |
| 117 | 2024 | [Hybrid Text Retrieval with Large Language Models: A Study of Robustness and Generalization](https://aclanthology.org/2024.lrec-main.748/) | LREC-COLING 2024; peer-reviewed | `reranking`, `hybrid`, `robustness` |
| 118 | 2024 | [Introducing Contextual Retrieval](https://www.anthropic.com/engineering/contextual-retrieval) | Anthropic Engineering; industry-report | `contextual-retrieval`, `chunking`, `hybrid` |
| 119 | 2024 | [LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression](https://aclanthology.org/2024.findings-acl.57/) | Findings ACL 2024; peer-reviewed | `prompt-compression`, `distillation` |
| 120 | 2024 | [Long-Context LLMs Meet RAG: Overcoming Challenges for Long Inputs in RAG](https://proceedings.iclr.cc/paper_files/paper/2025/hash/5df56b0238e56b953d4600d1b9e83982-Abstract-Conference.html) | ICLR 2025; peer-reviewed | `long-context`, `distraction`, `top-k` |
| 121 | 2024 | [LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory](https://openreview.net/forum?id=pZiyCaVuti) | ICLR 2025; peer-reviewed | `memory`, `benchmark`, `updates` |
| 122 | 2024 | [MMTEB: Massive Multilingual Text Embedding Benchmark](https://proceedings.iclr.cc/paper_files/paper/2025/file/fc0e3f908a2116ba529ad0a1530a3675-Paper-Conference.pdf) | ICLR 2025; peer-reviewed | `multilingual`, `embeddings`, `benchmark` |
| 123 | 2024 | [NoMIRACL: Knowing When You Don't Know for Robust Multilingual Retrieval-Augmented Generation](https://aclanthology.org/2024.findings-emnlp.730/) | Findings EMNLP 2024; peer-reviewed | `multilingual`, `abstention`, `hallucination` |
| 124 | 2024 | [PDF-to-Tree: Parsing PDF Content into a Tree Structure](https://aclanthology.org/2024.findings-emnlp.628/) | Findings EMNLP 2024; peer-reviewed | `pdf`, `parsing`, `hierarchy` |
| 125 | 2024 | [PoisonedRAG: Knowledge Poisoning Attacks to Retrieval-Augmented Generation of Large Language Models](https://www.usenix.org/conference/usenixsecurity25/presentation/zou-poisonedrag) | USENIX Security 2025; peer-reviewed | `security`, `poisoning` |
| 126 | 2024 | [RAGBench: Explainable Benchmark for Retrieval-Augmented Generation Systems](https://arxiv.org/abs/2407.11005) | arXiv; preprint | `benchmark`, `evaluation`, `trace` |
| 127 | 2024 | [RAGChecker: A Fine-Grained Framework for Diagnosing Retrieval-Augmented Generation](https://proceedings.neurips.cc/paper_files/paper/2024/file/27245589131d17368cccdfa990cbf16e-Paper-Datasets_and_Benchmarks_Track.pdf) | NeurIPS 2024 Datasets and Benchmarks; peer-reviewed | `evaluation`, `claims`, `diagnostics` |
| 128 | 2024 | [RaLMSpec: Accelerating Retrieval-Augmented Language Model Serving with Speculation](https://proceedings.mlr.press/v235/zhang24cq.html) | ICML 2024; peer-reviewed | `systems`, `latency`, `speculation` |
| 129 | 2024 | [RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs](https://proceedings.neurips.cc/paper_files/paper/2024/hash/db93ccb7cb70e655c1af7d7a2433e6ae-Abstract-Conference.html) | NeurIPS 2024; peer-reviewed | `reranking`, `instruction-tuning` |
| 130 | 2024 | [Sufficient Context: A New Lens on Retrieval Augmented Generation Systems](https://openreview.net/forum?id=8N8hWwTj6D) | ICLR 2025; peer-reviewed | `abstention`, `context-sufficiency` |
| 131 | 2024 | [The Good and The Bad: Exploring Privacy Issues in Retrieval-Augmented Generation](https://aclanthology.org/2024.findings-acl.267/) | Findings ACL 2024; peer-reviewed | `privacy`, `extraction`, `rag` |
| 132 | 2024 | [TREC Retrieval-Augmented Generation Track](https://trec-rag.github.io/) | NIST TREC 2024-2026; benchmark-program | `benchmark`, `citations`, `external-judgments` |
| 133 | 2024-01 | [MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries](https://arxiv.org/abs/2401.15391) | arXiv; preprint | `multi-hop`, `rag`, `benchmark` |
| 134 | 2024-01-29 | [Corrective Retrieval Augmented Generation](https://arxiv.org/abs/2401.15884) | arXiv; preprint | `correction`, `web-search`, `routing` |
| 135 | 2024-01-30 | [CRUD-RAG: A Comprehensive Chinese Benchmark for Retrieval-Augmented Generation](https://doi.org/10.1145/3701228) | ACM TOIS; peer-reviewed | `evaluation`, `crud`, `chinese` |
| 136 | 2024-01-31 | [RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval](https://openreview.net/forum?id=GN921JHCRw) | ICLR 2024; peer-reviewed | `hierarchical-retrieval`, `summaries` |
| 137 | 2024-03 | [MedRAG: Enhancing Large Language Models in Medicine with Retrieval-Augmented Generation](https://arxiv.org/abs/2403.04115) | arXiv; preprint | `medical`, `rag`, `domain` |
| 138 | 2024-03 | [RAFT: Adapting Language Model to Domain Specific RAG](https://arxiv.org/abs/2403.10131) | arXiv; preprint | `domain-adaptation`, `fine-tuning`, `rag` |
| 139 | 2024-03 | [Repoformer: Selective Retrieval for Repository-Level Code Completion](https://proceedings.mlr.press/v235/wu24a.html) | ICML 2024; peer-reviewed | `code`, `selective-retrieval`, `generation` |
| 140 | 2024-03-21 | [Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity](https://aclanthology.org/2024.naacl-long.389/) | NAACL 2024; peer-reviewed | `routing`, `complexity`, `efficiency` |
| 141 | 2024-04-24 | [From Local to Global: A Graph RAG Approach to Query-Focused Summarization](https://www.microsoft.com/en-us/research/publication/from-local-to-global-a-graph-rag-approach-to-query-focused-summarization/) | Microsoft Research; industry-report | `graph`, `global-synthesis`, `communities` |
| 142 | 2024-06-27 | [ColPali: Efficient Document Retrieval with Vision Language Models](https://proceedings.iclr.cc/paper_files/paper/2025/hash/99e9cf99cc114c46c2e6168e4dc0c43a-Abstract-Conference.html) | ICLR 2025; peer-reviewed | `visual-retrieval`, `late-interaction`, `documents` |
| 143 | 2024-07 | [BRIGHT: A Realistic and Challenging Benchmark for Reasoning-Intensive Retrieval](https://proceedings.iclr.cc/paper_files/paper/2025/file/7a0f8055c838df8e62329a76c7c6403d-Paper-Conference.pdf) | ICLR 2025; peer-reviewed | `benchmark`, `reasoning-retrieval` |
| 144 | 2024-07-23 | [Retrieval Augmented Generation or Long-Context LLMs? A Comprehensive Study and Hybrid Approach](https://aclanthology.org/2024.emnlp-industry.66/) | EMNLP Industry 2024; peer-reviewed | `long-context`, `routing`, `self-route` |
| 145 | 2024-09 | [Late Chunking: Contextual Chunk Embeddings Using Long-Context Embedding Models](https://arxiv.org/abs/2409.04701) | arXiv; preprint | `chunking`, `contextual-embeddings` |
| 146 | 2024-10-14 | [VisRAG: Vision-based Retrieval-Augmented Generation on Multi-modality Documents](https://proceedings.iclr.cc/paper_files/paper/2025/hash/3640e20b253c7530cce06abdd3c2361b-Abstract-Conference.html) | ICLR 2025; peer-reviewed | `visual-rag`, `documents`, `generation` |
### 2025

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 147 | 2025 | [Auditing Prompt Caching in Language Model APIs](https://proceedings.mlr.press/v267/gu25b.html) | ICML 2025; peer-reviewed | `caching`, `privacy`, `side-channel` |
| 148 | 2025 | [Beyond Text: Unveiling Privacy Vulnerabilities in Multi-modal Retrieval-Augmented Generation](https://aclanthology.org/2025.emnlp-main.1259/) | EMNLP 2025; peer-reviewed | `privacy`, `multimodal`, `extraction` |
| 149 | 2025 | [ComRAG: A Conversational Retrieval-Augmented Generation Framework with Dynamic Memory Consolidation](https://aclanthology.org/2025.acl-industry.53/) | ACL Industry 2025; peer-reviewed | `memory`, `consolidation`, `latency` |
| 150 | 2025 | [DeepRAG: Thinking to Retrieval Step by Step for Large Language Models](https://iclr.cc/virtual/2026/poster/10009155) | ICLR 2026; peer-reviewed | `retrieval-policy`, `mdp`, `reasoning` |
| 151 | 2025 | [DioR: Adaptive Cognitive Detection and Contextual Retrieval Optimization for Dynamic Retrieval-Augmented Generation](https://aclanthology.org/2025.acl-long.148/) | ACL 2025; peer-reviewed | `dynamic-rag`, `retrieval-trigger`, `context-selection` |
| 152 | 2025 | [From RAG to Memory: Non-Parametric Continual Learning for Large Language Models](https://proceedings.mlr.press/v267/gutierrez25a.html) | ICML 2025; peer-reviewed | `graph`, `memory`, `multi-hop` |
| 153 | 2025 | [Frustratingly Simple Retrieval Improves Challenging, Reasoning-Intensive Benchmarks](https://iclr.cc/virtual/2026/poster/10011084) | ICLR 2026; peer-reviewed | `datastore`, `systems`, `reasoning` |
| 154 | 2025 | [GaRAGe: A Benchmark for Grounded and Reliable RAG Evaluation](https://aclanthology.org/2025.findings-acl.875/) | Findings ACL 2025; peer-reviewed | `grounding`, `deflection`, `citations` |
| 155 | 2025 | [HiPRAG: Hierarchical Process Rewards for Retrieval-Augmented Generation](https://iclr.cc/virtual/2026/poster/10010451) | ICLR 2026; peer-reviewed | `process-reward`, `search-efficiency` |
| 156 | 2025 | [How Does Knowledge Selection Help Retrieval Augmented Generation?](https://aclanthology.org/2025.findings-emnlp.218/) | Findings EMNLP 2025; peer-reviewed | `selection`, `reranking`, `generation` |
| 157 | 2025 | [Knowledgeable-R1: Learning to Know When to Search and Trust External Knowledge](https://iclr.cc/virtual/2026/poster/10011379) | ICLR 2026; peer-reviewed | `conflict`, `parametric-knowledge`, `reinforcement-learning` |
| 158 | 2025 | [Learning Contextual Retrieval for Robust Conversational Search](https://aclanthology.org/2025.emnlp-main.602/) | EMNLP 2025; peer-reviewed | `conversation`, `contextual-retrieval`, `embeddings` |
| 159 | 2025 | [Learning Distraction-Aware Retrieval for Retrieval-Augmented Generation](https://iclr.cc/virtual/2026/poster/10008538) | ICLR 2026; peer-reviewed | `distraction`, `evidence-utility`, `long-context` |
| 160 | 2025 | [M+: Extending MemoryLLM with Scalable Long-Term Memory](https://proceedings.mlr.press/v267/wang25au.html) | ICML 2025; peer-reviewed | `latent-memory`, `retrieval` |
| 161 | 2025 | [M3DocVQA: A Benchmark for Multi-Modal Multi-Document Question Answering](https://openaccess.thecvf.com/content/ICCV2025W/MIRU/html/Cho_M3DocVQA_A_Benchmark_for_Multi-Modal_Multi-Document_Question_Answering_ICCVW_2025_paper.html) | ICCV 2025 Workshop; peer-reviewed | `multimodal`, `multi-document`, `benchmark` |
| 162 | 2025 | [MoLoRAG: Bootstrapping VLM-Based Retrieval with a Multi-Modal Document Graph](https://aclanthology.org/2025.emnlp-main.708/) | EMNLP 2025; peer-reviewed | `visual-rag`, `graph`, `documents` |
| 163 | 2025 | [mt RAG: A Multi-Turn Conversational Benchmark for Evaluating Retrieval-Augmented Generation Systems](https://aclanthology.org/2025.tacl-1.36/) | TACL 2025; peer-reviewed | `multi-turn`, `benchmark` |
| 164 | 2025 | [PropRAG: Guiding Retrieval with Beam Search over Proposition Paths](https://aclanthology.org/2025.emnlp-main.317/) | EMNLP 2025; peer-reviewed | `propositions`, `beam-search`, `multi-hop` |
| 165 | 2025 | [Q-RAG: Learning to Select Evidence with Value-Based Reinforcement Learning](https://iclr.cc/virtual/2026/poster/10009944) | ICLR 2026 Oral; peer-reviewed | `retrieval-policy`, `value-learning`, `long-context` |
| 166 | 2025 | [RAG LLMs Are Not Safer: A Safety Analysis of Retrieval-Augmented Generation for Large Language Models](https://aclanthology.org/2025.naacl-long.281/) | NAACL 2025; peer-reviewed | `safety`, `robustness` |
| 167 | 2025 | [REAL-MM-RAG: A Real-World Multi-Modal Retrieval Augmented Generation Benchmark](https://aclanthology.org/2025.acl-long.1528/) | ACL 2025; peer-reviewed | `multimodal`, `benchmark` |
| 168 | 2025 | [ReasonIR: Training Retrievers for Reasoning Tasks](https://openreview.net/forum?id=kkBCNLMbGj) | COLM 2025; peer-reviewed | `reasoning-retrieval`, `hard-negatives` |
| 169 | 2025 | [RemoteRAG: A Privacy-Preserving LLM Cloud RAG Service](https://aclanthology.org/2025.findings-acl.197/) | Findings ACL 2025; peer-reviewed | `privacy`, `differential-privacy`, `cloud` |
| 170 | 2025 | [ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning](https://openreview.net/forum?id=OuGAwwAT8G) | NeurIPS 2025; peer-reviewed | `agentic-rag`, `grpo` |
| 171 | 2025 | [Retrieval-Augmented Reasoning with Query-Specific Knowledge Graphs](https://iclr.cc/virtual/2026/poster/10008199) | ICLR 2026; peer-reviewed | `dynamic-graph`, `reasoning` |
| 172 | 2025 | [RMM: Reinforced Memory Management for Long-Term Conversational Agents](https://aclanthology.org/2025.acl-long.413/) | ACL 2025; peer-reviewed | `memory`, `conversation`, `reinforcement-learning` |
| 173 | 2025 | [SafeRAG: Benchmarking Security in Retrieval-Augmented Generation of Large Language Model](https://aclanthology.org/2025.acl-long.230/) | ACL 2025; peer-reviewed | `security`, `benchmark`, `conflict` |
| 174 | 2025 | [Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning](https://openreview.net/forum?id=Rwhi91ideu) | COLM 2025; peer-reviewed | `agentic-rag`, `reinforcement-learning`, `search` |
| 175 | 2025 | [SeCon-RAG: A Security-Conscious Retrieval-Augmented Generation Framework](https://proceedings.neurips.cc/paper_files/paper/2025/hash/668563ef18fbfef0b66af491ea334d5f-Abstract-Conference.html) | NeurIPS 2025; peer-reviewed | `security`, `filtering`, `conflict` |
| 176 | 2025 | [Shifting from Ranking to Set Selection for Retrieval Augmented Generation](https://aclanthology.org/2025.acl-long.861/) | ACL 2025; peer-reviewed | `set-selection`, `multi-hop`, `reranking` |
| 177 | 2025 | [StepSearch: Igniting LLMs Search Ability via Step-Wise Proximal Policy Optimization](https://aclanthology.org/2025.emnlp-main.1106/) | EMNLP 2025; peer-reviewed | `agentic-rag`, `process-reward` |
| 178 | 2025 | [Stronger Baselines for Retrieval-Augmented Generation with Long-Context Language Models](https://aclanthology.org/2025.emnlp-main.1656/) | EMNLP 2025; peer-reviewed | `long-context`, `rag`, `baselines` |
| 179 | 2025 | [syftr: Pareto-Optimal Generative AI](https://proceedings.mlr.press/v293/conway25a.html) | AutoML / PMLR 2025; peer-reviewed | `automl`, `pareto`, `cost` |
| 180 | 2025 | [TableRAG: A Retrieval Augmented Generation Framework for Heterogeneous Document Reasoning](https://aclanthology.org/2025.emnlp-main.710/) | EMNLP 2025; peer-reviewed | `tables`, `sql`, `heterogeneous-documents` |
| 181 | 2025 | [Think&Cite: Improving Attributed Text Generation with Self-Guided MCTS](https://aclanthology.org/2025.acl-long.490/) | ACL 2025; peer-reviewed | `citations`, `mcts`, `attribution` |
| 182 | 2025 | [Visual Document Retrieval-Augmented Generation with Dynamic Token Compression](https://openaccess.thecvf.com/content/CVPR2025/html/Tanaka_Visual_Document_Retrieval-Augmented_Generation_with_Dynamic_Token_Compression_CVPR_2025_paper.html) | CVPR 2025; peer-reviewed | `visual-rag`, `compression` |
| 183 | 2025 | [When to Use Graphs in Retrieval-Augmented Generation](https://iclr.cc/virtual/2026/poster/10007992) | ICLR 2026; peer-reviewed | `graph`, `benchmark`, `component-analysis` |
### 2026

| # | First public | Work | Venue/status | Topics |
|---:|---|---|---|---|
| 184 | 2026 | [CODEPROMPTZIP: Code-specific Prompt Compression for Retrieval-Augmented Generation in Coding Tasks with LMs](https://aclanthology.org/2026.findings-acl.1384/) | Findings ACL 2026; peer-reviewed | `code`, `prompt-compression`, `rag` |
| 185 | 2026 | [Dissecting GraphRAG: A Modular Analysis of Knowledge Structuring for Factoid Question Answering](https://aclanthology.org/2026.tacl-1.29/) | TACL 2026; peer-reviewed | `graph-rag`, `ablation`, `evaluation` |
| 186 | 2026 | [Exposing Privacy Risks in Graph Retrieval-Augmented Generation](https://aclanthology.org/2026.findings-acl.899/) | Findings ACL 2026; peer-reviewed | `privacy`, `graph-rag`, `extraction` |
| 187 | 2026 | [HiChunk: Evaluating and Enhancing Retrieval Augmented Generation with Hierarchical Chunking](https://aclanthology.org/2026.acl-long.1372/) | ACL 2026; peer-reviewed | `chunking`, `hierarchical`, `evaluation` |
| 188 | 2026 | [MegaRAG: Multimodal Knowledge Graph Retrieval-Augmented Generation](https://aclanthology.org/2026.acl-long.2218/) | ACL 2026; peer-reviewed | `multimodal`, `graph` |
| 189 | 2026 | [NEST: Nested Evidence Survival for Retrieval](https://aclanthology.org/2026.acl-industry.35/) | ACL Industry 2026; peer-reviewed | `evidence-selection`, `noise`, `retrieval` |
| 190 | 2026 | [Overcoming the Retrieval Barrier: Indirect Prompt Injection in the Wild for LLM Systems](https://www.usenix.org/system/files/conference/usenixsecurity26/sec26_prepub_chang.pdf) | USENIX Security 2026; peer-reviewed | `security`, `prompt-injection`, `retrieval` |
| 191 | 2026 | [PRA-RAG: Provably Robust Aggregation for Retrieval-Augmented Generation](https://aclanthology.org/2026.findings-acl.1794/) | Findings ACL 2026; peer-reviewed | `security`, `provable-robustness`, `poisoning` |
| 192 | 2026 | [PROGRAM: Programmatic Retrieval Optimization with Generative Reasoning and Augmented Multi-queries](https://aclanthology.org/2026.findings-acl.1090/) | Findings ACL 2026; peer-reviewed | `programmatic-retrieval`, `multi-hop` |
| 193 | 2026 | [R3AG: Retriever Routing for Retrieval-Augmented Generation](https://aclanthology.org/2026.acl-long.939/) | ACL 2026; peer-reviewed | `routing`, `retrievers`, `rag` |
| 194 | 2026 | [RAG over Tables: Hierarchical Memory Index, Multi-Stage Retrieval, and Benchmarking](https://aclanthology.org/2026.findings-acl.1902/) | Findings ACL 2026; peer-reviewed | `tables`, `hierarchical-index`, `graph` |
| 195 | 2026 | [Region-R1: Reinforcing Query-Side Region Cropping for Multi-Modal Re-Ranking](https://aclanthology.org/2026.findings-acl.510/) | Findings ACL 2026; peer-reviewed | `multimodal`, `reranking`, `reinforcement-learning` |
| 196 | 2026 | [Retrieval as Generation: A Unified Framework with Self-Triggered Information Planning](https://aclanthology.org/2026.acl-long.196/) | ACL 2026; peer-reviewed | `retrieval-control`, `generation`, `reinforcement-learning` |
| 197 | 2026 | [RobustVisRAG: Robust Retrieval-Augmented Generation for Real-World Visual Document Understanding](https://openaccess.thecvf.com/content/CVPR2026/html/Zhang_RobustVisRAG_Robust_Retrieval-Augmented_Generation_for_Real-World_Visual_Document_Understanding_CVPR_2026_paper.html) | CVPR 2026; peer-reviewed | `visual-rag`, `robustness`, `distortion` |
| 198 | 2026 | [RouteRAG: Efficient Retrieval-Augmented Generation from Text and Graph via Reinforcement Learning](https://aclanthology.org/2026.findings-acl.1502/) | Findings ACL 2026; peer-reviewed | `routing`, `graph`, `reinforcement-learning` |
| 199 | 2026 | [SCAN: Semantic Document Layout Analysis for Textual and Visual Retrieval-Augmented Generation](https://aclanthology.org/2026.findings-eacl.82/) | Findings EACL 2026; peer-reviewed | `layout`, `visual-rag`, `document-parsing` |
| 200 | 2026 | [SemEval-2026 Task 8: MTRAGEval - Evaluating Multi-Turn Retrieval-Augmented Generation](https://aclanthology.org/2026.semeval-1.447/) | SemEval 2026; peer-reviewed | `multi-turn`, `benchmark`, `retrieval` |
| 201 | 2026 | [T2-RAGBench: Text-and-Table Benchmark for Evaluating Retrieval-Augmented Generation](https://aclanthology.org/2026.eacl-long.8/) | EACL 2026; peer-reviewed | `tables`, `benchmark`, `numerical-reasoning` |
| 202 | 2026 | [Tackling Distractor Documents in Multi-Hop QA with Reinforcement and Curriculum Learning](https://aclanthology.org/2026.findings-eacl.294/) | Findings EACL 2026; peer-reviewed | `citations`, `curriculum`, `distractors` |
| 203 | 2026 | [When Good OCR Is Not Enough: Benchmarking OCR Robustness for Retrieval-Augmented Generation](https://aclanthology.org/2026.acl-industry.60/) | ACL Industry 2026; peer-reviewed | `ocr`, `benchmark`, `document-rag` |

## Coverage and maintenance rules

The index is broad by design, but it is not a claim that every publication ever using retrieval appears here. A work enters the registry when it is needed to support a historical, mechanism, empirical, evaluation, security, or systems claim in this repository. The [coverage matrix](coverage_matrix.md) shows which lifecycle surface each body of work supports.

When adding a source:

1. prefer final official proceedings, then accepted-paper/author manuscript, then an original preprint or first-party report;
2. record earliest public date separately from venue year;
3. use a unique stable ID and primary URL;
4. label status without upgrading preprints or industry reports;
5. attach specific topic tags and update the substantive chapter;
6. regenerate this file and the complete handbook notebook; and
7. run the full validator so dates, links, notebook execution, and coverage remain synchronized.