Simplifications are Absolutists: How Simplified Language Reduces Word Sense Awareness in LLM-Generated Definitions
This addresses the problem of balancing simplicity and completeness in educational NLP for learners, though it appears incremental as it builds on existing simplification and fine-tuning approaches.
The study investigated how simplifying language in LLM-generated definitions affects word sense awareness, particularly for homonyms, finding that simplification drastically degrades definition completeness by neglecting polysemy and increasing misunderstanding risk, while fine-tuning with Direct Preference Optimization substantially improved homonym response quality across all prompt types.
Large Language Models (LLMs) can provide accurate word definitions and explanations for any context. However, the scope of the definition changes for different target groups, like children or language learners. This is especially relevant for homonyms, words with multiple meanings, where oversimplification might risk information loss by omitting key senses, potentially misleading users who trust LLM outputs. We investigate how simplification impacts homonym definition quality across three target groups: Normal, Simple, and ELI5. Using two novel evaluation datasets spanning multiple languages, we test DeepSeek v3, Llama 4 Maverick, Qwen3-30B A3B, GPT-4o mini, and Llama 3.1 8B via LLM-as-Judge and human annotations. Our results show that simplification drastically degrades definition completeness by neglecting polysemy, increasing the risk of misunderstanding. Fine-tuning Llama 3.1 8B with Direct Preference Optimization substantially improves homonym response quality across all prompt types. These findings highlight the need to balance simplicity and completeness in educational NLP to ensure reliable, context-aware definitions for all learners.