mlsanityData to 14 Oct 2026

DSL-R1: From SQL to DSL for Training Retrieval Agents across Structured and Unstructured Data with Reinforcement Learning

arXiv 2603.21018 ↗First posted 14 Jan 2026cs.IR cs.AI cs.DBcited by 0

On the map: Deep reinforcement learning methods → RL training of language and agent models

Abstract (arXiv)

Effective retrieval in complex domains requires bridging the gap between structured metadata and unstructured content. Existing systems typically isolate these capabilities, relying on either symbolic filtering or vector similarity, failing to capture their interplay. In this work, we propose DSL-R1, a unified framework that synergizes logical reasoning with semantic matching via a novel Domain-Specific Language (DSL). By embedding vector primitives within SQL-style operators, our approach leverages the complementary strengths of symbolic precision and semantic coverage. We further introduce a reinforcement learning mechanism where rule-based execution feedback and retrieval quality rewards jointly optimize the DSL generation, balancing structural correctness and semantic alignment. Evaluations on a large-scale industrial email benchmark demonstrate that DSL-R1 achieves a +12.3% improvement in Hit@1/3, consistently outperforming decoupled baselines and establishing a robust paradigm for hybrid retrieval.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods. Builds on 12 of its 24 references that are in the map. First posted on arXiv 14 Jan 2026.

Builds on

12 of 24 references.

References · 12

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  3. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,864
  4. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  5. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,876
  6. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 2,057
  7. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,210
  8. 2021Show Your Work: Scratchpads for Intermediate Computation with Language Modelscited by 886
  9. 2020Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrievalcited by 849
  10. 2017Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learningcited by 687
  11. 2016DeepCoder: Learning to Write Programscited by 149
  12. 2024STaRK: Benchmarking LLM Retrieval on Textual and Relational Knowledge Basescited by 55

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026RL-Index: Reinforcement Learning for Retrieval Index Reasoning
  2. 2026RedParrot: Accelerating NL-to-DSL for Business Analytics via Query Semantic Caching
  3. 2025Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL
  4. 2026ProRetrieval: Learning to Orchestrate Hybrid Search via Executable Program Synthesis ↗
  5. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
  6. 2025ParallelSearch: Train your LLMs to Decompose Query and Search Sub-queries in Parallel with Reinforcement Learning
  7. 2025CogniSQL-R1-Zero: Lightweight Reinforced Reasoning for Efficient SQL Generation
  8. 2025R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning