TaxoLLaMA: WordNet-based Model for Solving Multiple Lexical Semantic Tasks
This work addresses lexical semantic tasks for NLP researchers, but it is incremental as it applies existing methods (quantization and LoRA) to a new model.
The authors tackled the problem of capturing lexical-semantic knowledge from WordNet using LLMs, resulting in TaxoLLaMA, a lightweight model that achieves 11 state-of-the-art results and 4 top-2 results across 16 tasks in taxonomy-related tasks, with strong zero-shot performance.
In this paper, we explore the capabilities of LLMs in capturing lexical-semantic knowledge from WordNet on the example of the LLaMA-2-7b model and test it on multiple lexical semantic tasks. As the outcome of our experiments, we present TaxoLLaMA, the everything-in-one model, lightweight due to 4-bit quantization and LoRA. It achieves 11 SotA results, 4 top-2 results out of 16 tasks for the Taxonomy Enrichment, Hypernym Discovery, Taxonomy Construction, and Lexical Entailment tasks. Moreover, it demonstrates very strong zero-shot performance on Lexical Entailment and Taxonomy Construction with no fine-tuning. We also explore its hidden multilingual and domain adaptation capabilities with a little tuning or few-shot learning. All datasets, code, and model are available online at https://github.com/VityaVitalich/TaxoLLaMA