aiqbench analysis
KBQA-R1 addresses critical limitations in Knowledge Base Question Answering by shifting from text imitation to interaction optimization via Reinforcement Learning. By leveraging Group Relative Policy Optimization (GRPO) and Referenced Rejection Sampling, the framework grounds LLM reasoning in verifiable execution rather than relying on static supervision or hallucinated queries. This approach effectively resolves cold-start challenges and achieves state-of-the-art performance on standard benchmarks by treating KBQA as a multi-turn decision process.
KBQA-R1: Reinforcing Large Language Models for Knowledge Base Question Answering
arXiv:2512.10999v3 Announce Type: replace Abstract: Knowledge Base Question Answering (KBQA) challenges models to bridge the gap between natural language and strict knowledge graph schemas by generating executable logical forms. While Large Language Models (LLMs) have advanced this field, current approaches often struggle with a dichotomy of failure: they either generate hallucinated queries without verifying schema existence or exhibit rigid, template-based reasoning that mimics synthesized traces without true comprehension of the environment. To address these limitations, we present \textbf{KBQA-R1}, a framework that shifts the paradigm from text imitation to interaction optimization via Reinforcement Learning. Treating KBQA as a multi-turn decision process, our model learns to navigate the knowledge base using a list of actions, leveraging Group Relative Policy Optimization (GRPO) to refine its strategies based on concrete execution feedback rather than static supervision. Furthermore, we introduce \textbf{Referenced Rejection Sampling (RRS)}, a data synthesis method that resolves cold-start challenges by strictly aligning reasoning traces with ground-truth action sequences. Extensive experiments on WebQSP, GrailQA, and GraphQuestions demonstrate that KBQA-R1 achieves state-of-the-art performance, effectively grounding LLM reasoning in verifiable execution.
Abstract:Knowledge Base Question Answering (KBQA) challenges models to bridge the gap between natural language and strict knowledge graph schemas by generating executable logical forms. While Large Language Models (LLMs) have advanced this field, current approaches often struggle with a dichotomy of failure: they either generate hallucinated queries without verifying schema existence or exhibit rigid, template-based reasoning that mimics synthesized traces without true comprehension of the environment. To address these limitations, we present \textbf{KBQA-R1}, a framework that shifts the paradigm from text imitation to interaction optimization via Reinforcement Learning. Treating KBQA as a multi-turn decision process, our model learns to navigate the knowledge base using a list of actions, leveraging Group Relative Policy Optimization (GRPO) to refine its strategies based on concrete execution feedback rather than static supervision. Furthermore, we introduce \textbf{Referenced Rejection Sampling (RRS)}, a data synthesis method that resolves cold-start challenges by strictly aligning reasoning traces with ground-truth action sequences. Extensive experiments on WebQSP, GrailQA, and GraphQuestions demonstrate that KBQA-R1 achieves state-of-the-art performance, effectively grounding LLM reasoning in verifiable execution.
Computation and Language (cs.CL)
arXiv:2512.10999 [cs.CL]
(or arXiv:2512.10999v3 [cs.CL] for this version)
Full text under the publisher's open licence · 183 words
Related coverage
Models Can't Remember Their Training. Neither Can You.
LessWrong · Jul 19, 2026
Opaque recurrence, and other AI terms that you should probably know
TechCrunch · Sep 7, 2026
Salesforce after Dreamforce: How the CRM giant can grow beyond its own interface
SiliconANGLE · Sep 19, 2026
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026