Reproduce “KBQA-R1: Reinforcing Large Language Models for Knowledge Base Question Answering” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Xin Sun, Zhongqi Chen, Xing Zheng, Qiang Liu, Shu Wu, Bowen Song, Zilei Wang, Weiqiang Wang and 1 more
KBQA-R1 answers knowledge-base questions by building queries through executable actions, observing results, and using reinforcement learning to train action selection. Its multi-turn process validates relations against the schema and updates later steps from KB feedback. On GrailQA’s zero-shot dev split, it scores 85.2 ± 0.4 entity-level F1, versus 76.1 ± 0.4 for KBQA-o1; this is a result on Freebase, not arbitrary knowledge bases.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.