Learning to Retrieve via Reinforcement Learning in Embedding Space

arXiv:2610.07731 · cs.IR, cs.AI, cs.CL, cs.LG · Submitted 2026-10-06 · Read on arXiv

cs.IR, cs.AI, cs.CL, cs.LG

Submitted: 2026-10-06

Updated: 2026-10-06

Code: https://github.com/liuqi6777/RELER

Project page: https://wouterkool.github.io/publication/buy-4-samples-free-baseline

Terminology

• Reinforcement Learning — 1× in this paper · explained in 4 episodes

Sources

• Back to Basics: Revisiting REINFORCE Style Optimization for Learning from Human Feedback in LLMs
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• Embed-RL: Reinforcement Learning for Reasoning-Driven Multimodal Embeddings
• Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
• Product-form estimators: exploiting independence to scale up Monte Carlo
• Towards General Text Embeddings with Multi-stage Contrastive Learning
• E2Rank: Unifying Text Embedding and Listwise Reranking for Effective and Efficient Search
• ReasonRank: Empowering Passage Ranking with Strong Reasoning Ability
• Retrieval-GRPO: A Multi-Objective Reinforcement Learning Framework for Dense Retrieval in Taobao Search
• DIVER: A Multi-Stage Approach for Reasoning-intensive Information Retrieval
• MemGPT: Towards LLMs as Operating Systems
• NeuralNDCG: Direct Optimisation of a Ranking Metric via Differentiable Relaxation of Sorting
• ReasonIR: Training Retrievers for Reasoning Tasks
• Representation Learning with Contrastive Predictive Coding
• Variance Reduction in Gradient Exploration for Online Learning to Rank
• Text Embeddings by Weakly-Supervised Contrastive Pre-training
• Learning To Retrieve: How to Train a Dense Retrieval Model Effectively and Efficiently
• Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models
• Reinforcement Fine-Tuning for History-Aware Dense Retriever in RAG
• Rank-R1: Enhancing Reasoning in LLM-based Document Rerankers via Reinforcement Learning

Related papers

• The Price of Isolation: Estimating the Ecosystem Cost of Symmetric Two-Sided A/B Testing
• SCAR: Semantic Continuity-Aware Retrieval for Efficient Context Expansion in RAG
• MixLoRA-DSI: Dynamically Expandable Mixture-of-LoRA Experts for Rehearsal-Free Generative Retrieval over Dynamic Corpora
• How Do LLM-Generated Texts Impact Term-Based Retrieval Models?
• MemSyco-Bench: Benchmarking Sycophancy in Agent Memory
• Training with Missed Targets in Generative Recommendation: Separating Supervision from Probability Competition