mlsanityData to 7 Oct 2026

Zero-Shot Retrieval with Search Agents and Hybrid Environments

arXiv 2209.15469 ↗First posted 30 Sep 2022cs.CLcited by 6

On the map: LLM agents and tool use → Web, search, and research agents

Abstract (arXiv)

Learning to search is the task of building artificial agents that learn to autonomously use a search box to find information. So far, it has been shown that current language models can learn symbolic query reformulation policies, in combination with traditional term-based retrieval, but fall short of outperforming neural retrievers. We extend the previous learning to search setup to a hybrid environment, which accepts discrete query refinement operations, after a first-pass retrieval step via a dual encoder. Experiments on the BEIR task show that search agents, trained via behavioral cloning, outperform the underlying search system based on a combined dual encoder retriever and cross encoder reranker. Furthermore, we find that simple heuristic Hybrid Retrieval Environments (HRE) can improve baseline performance by several nDCG points. The search agent based on HRE (HARE) matches state-of-the-art performance, balanced in both zero-shot and in-domain evaluations, via interpretable actions, and at twice the speed.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use. Builds on 35 of its 49 references that are in the map. First posted on arXiv 30 Sep 2022.

Builds on

35 of 49 references.

References · 35

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  5. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  6. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,243
  7. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  8. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
  9. 2021WebGPT: Browser-assisted question-answering with human feedbackcited by 1,722
  10. 2016MS MARCO: A Human Generated MAchine Reading COmprehension Datasetcited by 1,591
  11. 2021BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Modelscited by 1,225
  12. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
23 more
  1. 2020ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERTcited by 1,160
  2. 2020Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrievalcited by 846
  3. 2019Passage Re-ranking with BERTcited by 629
  4. 2021ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interactioncited by 464
  5. 2021Large Dual Encoders Are Generalizable Retrieverscited by 454
  6. 2020Document Ranking with a Pretrained Sequence-to-Sequence Modelcited by 373
  7. 2021SPLADE: Sparse Lexical and Expansion Model for First Stage Rankingcited by 283
  8. 2022SGPT: GPT Sentence Embeddings for Semantic Searchcited by 204
  9. 2021Simple Entity-Centric Questions Challenge Dense Retrieverscited by 140
  10. 2020Question and Answer Test-Train Overlap in Open-Domain Question Answering Datasetscited by 137
  11. 2021SPLADE v2: Sparse Lexical and Expansion Model for Information Retrievalcited by 126
  12. 2014Asymmetric LSH (ALSH) for Sublinear Time Maximum Inner Product Search (MIPS)cited by 109
  13. 2022From Distillation to Hard Negative Sampling: Making Sparse Neural IR Models More Effectivecited by 98
  14. 2019Answering Complex Open-domain Questions Through Iterative Query Generationcited by 60
  15. 2021Salient Phrase Aware Dense Retrieval: Can a Dense Retriever Imitate a Sparse One?cited by 59
  16. 2017Ask the Right Questions: Active Question Reformulation with Reinforcement Learningcited by 44
  17. 2020Answering Open-Domain Questions of Varying Reasoning Steps from Textcited by 43
  18. 2021Adaptive Information Seeking for Open-Domain Question Answeringcited by 29
  19. 2022Out-of-Domain Semantics to the Rescue! Zero-Shot Hybrid Retrieval Modelscited by 26
  20. 2022A Thorough Examination on Zero-shot Dense Retrievalcited by 21
  21. 2021Boosting Search Engines with Interactive Agentscited by 19
  22. 2022No Parameter Left Behind: How Distillation and Model Size Affect Zero-Shot Retrievalcited by 17
  23. 2013Query Expansion Using Term Distribution and Term Associationcited by 3

Built on by

Cited by year

Citing papers per year: 2022: 3; 2023: 1; 2024: 2.

6 papers.

  1. 2024Learning to Plan and Generate Text with Citationscited by 28
  2. 2022Visconde: Multi-document QA with GPT-3 and Neural Rerankingcited by 23
  3. 2024Robust Neural Information Retrieval: An Adversarial and Out-of-distribution Perspectivecited by 15
  4. 2023BERM: Training the Balanced and Extractable Representation for Matching to Improve Generalization Ability of Dense Retrievalcited by 11
  5. 2022Decoding a Neural Retriever's Latent Space for Query Suggestioncited by 5
  6. 2022Cheater's Bowl: Human vs. Computer Search Strategies for Open-Domain Question Answeringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025WebSeer: Training Deeper Search Agents through Reinforcement Learning with Self-Reflection
  2. 2026Dr. Zero: Self-Evolving Search Agents without Training Data
  3. 2026Critic-R: Improving Agentic Search using Instruction-tuned Retrievers with Natural Language Introspective Feedback
  4. 2026ProRetrieval: Learning to Orchestrate Hybrid Search via Executable Program Synthesis ↗
  5. 2025MaskSearch: A Universal Pre-Training Framework to Enhance Agentic Search Capability
  6. 2026Agentic-R: Learning to Retrieve for Agentic Search
  7. 2026A Reference Architecture for Agentic Hybrid Retrieval in Dataset Search
  8. 2026SE-Search: Self-Evolving Search Agent via Memory and Dense Reward