LLMs + Symbols: Picking the Right Meaning Without Labels
Words like "bank" can mean a riverbank or a financial institution. Getting the right meaning—called word sense disambiguation—is hard for AI, especially when we need rich, logic-friendly knowledge for reasoning.
This paper introduces an annotation-free workflow that teams up symbolic language understanding with large language models:
- A symbolic system proposes multiple candidate meanings for a sentence.
- Each candidate is rewritten as a clear natural-language alternative.
- An LLM is asked which alternative fits the context best.
- The chosen meaning is sent back to the symbolic system for deeper inference.
Unlike methods tied to WordNet/FrameNet and hand-labeled data, this approach scales to richer knowledge bases (e.g., OpenCyc) without manual training. In tests against human gold standards, it selects appropriate meanings effectively.
Paper: https://arxiv.org/abs/2511.16577v1
Paper: https://arxiv.org/abs/2511.16577v1
Register: https://www.AiFeta.com
AI NLP LLM SymbolicAI WordSenseDisambiguation KnowledgeGraphs NLU
