The Effect of Model Size on LLM Post-hoc Explainability via LIME

arXiv:2405.05348v1 Announce Type: cross
Abstract: Large language models (LLMs) are becoming bigger to boost performance. However, little is known about how explainability is affected by this trend. This work explores LIME explanations for DeBERTaV3 models of four different sizes on natural language inference (NLI) and zero-shot classification (ZSC) tasks. We evaluate the explanations based on their faithfulness to the models' internal decision processes and their plausibility, i.e. their agreement with human explanations. The key finding is that increased model size …

abstract arxiv bigger boost classification cs.ai cs.cl cs.lg explainability however inference language language models large language large language models lime llm llms natural natural language performance tasks trend type via work zero-shot