Search papers, labs, and topics across Lattice.
The paper introduces NeuroSymActive, a neural-symbolic framework for Knowledge Graph Question Answering (KGQA) that combines differentiable reasoning with active exploration. It addresses the limitations of both naive graph embedding and purely symbolic approaches by using a soft-unification style symbolic module coupled with a neural path evaluator and a Monte-Carlo tree search-based exploration policy. Experiments on KGQA benchmarks demonstrate that NeuroSymActive achieves high answer accuracy with fewer graph lookups and model calls compared to retrieval-augmented baselines.
By actively exploring knowledge graphs with a differentiable neural-symbolic approach, NeuroSymActive achieves strong KGQA accuracy while drastically reducing the computational cost of graph lookups.
Large pretrained language models and neural reasoning systems have advanced many natural language tasks, yet they remain challenged by knowledge-intensive queries that require precise, structured multi-hop inference. Knowledge graphs provide a compact symbolic substrate for factual grounding, but integrating graph structure with neural models is nontrivial: naively embedding graph facts into prompts leads to inefficiency and fragility, while purely symbolic or search-heavy approaches can be costly in retrievals and lack gradient-based refinement. We introduce NeuroSymActive, a modular framework that combines a differentiable neural-symbolic reasoning layer with an active, value-guided exploration controller for Knowledge Graph Question Answering. The method couples soft-unification style symbolic modules with a neural path evaluator and a Monte-Carlo style exploration policy that prioritizes high-value path expansions. Empirical results on standard KGQA benchmarks show that NeuroSymActive attains strong answer accuracy while reducing the number of expensive graph lookups and model calls compared to common retrieval-augmented baselines.
