# Text-to-SPARQL Generation with Reinforcement Learning: A GRPO-based Approach on DBLP

arXiv: https://arxiv.org/abs/2605.20066 · HTML: https://mlsanity.com/paper/2605.20066/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 19 of its 34 references that are in the map.
- First posted on arXiv 19 May 2026.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)

## Builds on (most cited first, 19 of 19)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [DoRA: Weight-Decomposed Low-Rank Adaptation](https://mlsanity.com/paper/2402.09353/) (2024)
- [Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs](https://mlsanity.com/paper/2503.01307/) (2025)
- [Spurious Rewards: Rethinking Training Signals in RLVR](https://mlsanity.com/paper/2506.10947/) (2025)
- [OctoThinker: Mid-training Incentivizes Reinforcement Learning Scaling](https://mlsanity.com/paper/2506.20512/) (2025)
- [KG-Agent: An Efficient Autonomous Agent Framework for Complex Reasoning over Knowledge Graph](https://mlsanity.com/paper/2402.11163/) (2024)
- [Domain-specific Knowledge Graphs: A survey](https://mlsanity.com/paper/2011.00235/) (2020)
- [Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQL](https://mlsanity.com/paper/2503.23157/) (2025)
- [QALD-9-plus: A Multilingual Dataset for Question Answering over DBpedia and Wikidata Translated by Native Speakers](https://mlsanity.com/paper/2202.00120/) (2022)
- [Modern Baselines for SPARQL Semantic Parsing](https://mlsanity.com/paper/2204.12793/) (2022)
- [DBLP-QuAD: A Question Answering Dataset over the DBLP Scholarly Knowledge Graph](https://mlsanity.com/paper/2303.13351/) (2023)
- [Sparks of Tabular Reasoning via Text2SQL Reinforcement Learning](https://mlsanity.com/paper/2505.00016/) (2025)
- [Question Answering on Scholarly Knowledge Graphs](https://mlsanity.com/paper/2006.01527/) (2020)
- [Efficient and Transferable Agentic Knowledge Graph RAG via Reinforcement Learning](https://mlsanity.com/paper/2509.26383/) (2025)
- [A Copy Mechanism for Handling Knowledge Base Elements in SPARQL Neural Machine Translation](https://mlsanity.com/paper/2211.10271/) (2022)
- [PSYCHIC: A Neuro-Symbolic Framework for Knowledge Graph Question-Answering Grounding](https://mlsanity.com/paper/2310.12638/) (2023)
- [KnowCoder-A1: Incentivizing Agentic Reasoning Capability with Outcome Supervision for KBQA](https://mlsanity.com/paper/2510.25101/) (2025)
- [Learning to Refine: An Agentic RL Approach for Iterative SPARQL Query Construction](https://mlsanity.com/paper/2511.11770/) (2025)
- [ConstrainedSQL: Training LLMs for Text2SQL via Constrained Reinforcement Learning](https://mlsanity.com/paper/2511.09693/) (2025)

## Built on by (most cited first, 1 of 1)

- [Enabling Knowledge Graph Understanding at Scale with the EXplore Your Graphs ENgine (EXYGEN)](https://mlsanity.com/paper/2609.11569/) (2026)

## Abstract (arXiv)

Knowledge graph question answering seeks to translate natural language questions into executable queries over knowledge graphs, but existing approaches often rely on large models or full supervision in the form of gold query annotations. This study examines whether reinforcement learning with outcome-based rewards can train a small instruction-tuned language model to perform zero-shot Text-to-SPARQL generation in the scholarly domain. Group-Relative Policy Optimization (GRPO) is applied to the Qwen3-1.7B model on DBLP-QuAD, using prompts that combine natural language questions with symbolic hints about entities and relations. Training relies on execution feedback, structural constraints, and answer-level rewards, with an additional variant that incorporates gold-query-based shaping. The resulting models are compared to the unmodified zero-shot baseline and to a supervised DoRA-finetuned baseline across answer-level accuracy, execution accuracy, category-wise scores, and generalization to held-out templates. GRPO substantially improves over the zero-shot baseline and exhibits competitive generalization, while supervised DoRA finetuning achieves higher overall accuracy on the same model scale. Ablation analyses indicate that execution-based rewards account for most gains, with additional shaping yielding limited additional benefit, suggesting that outcome-based reinforcement learning is a viable training strategy when gold queries are unavailable for token-level supervision.