Year: 2025
Type: conference-paper
Abstract: Natural Language Inference (NLI) relies heavily on adequately parsing the semantic content of the premise and hypothesis.In this work, we investigate whether adding semantic information in the form of an Abstract Meaning Representation (AMR) helps pretrained language models better generalize in NLI.Our experiments 1 integrating AMR into NLI in both fine-tuning and prompting settings show that the presence of AMR in fine-tuning hinders model generalization while prompting with AMR leads to slight gains in GPT-4o.However, an ablation study reveals that the improvement comes from amplifying surfacelevel differences rather than aiding semantic reasoning.This amplification can mislead models to predict non-entailment even when the core meaning is preserved.
Authors: Junghyun Min, Xiulin Yang, Shira Wein
Institutions: Amherst College, Georgetown University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Funder: Georgetown University
OpenAlex ID: W4416033808