MorphNLI: A Stepwise Approach to Natural Language Inference Using Text Morphing
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912902871515136 |
|---|---|
| author | Negru, Vlad Andrei Vacareanu, Robert Lemnaru, Camelia Surdeanu, Mihai Potolea, Rodica |
| author_facet | Negru, Vlad Andrei Vacareanu, Robert Lemnaru, Camelia Surdeanu, Mihai Potolea, Rodica |
| contents | We introduce MorphNLI, a modular step-by-step approach to natural language inference (NLI). When classifying the premise-hypothesis pairs into {entailment, contradiction, neutral}, we use a language model to generate the necessary edits to incrementally transform (i.e., morph) the premise into the hypothesis. Then, using an off-the-shelf NLI model we track how the entailment progresses with these atomic changes, aggregating these intermediate labels into a final output. We demonstrate the advantages of our proposed method particularly in realistic cross-domain settings, where our method always outperforms strong baselines with improvements up to 12.6% (relative). Further, our proposed approach is explainable as the atomic edits can be used to understand the overall NLI label. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_09567 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MorphNLI: A Stepwise Approach to Natural Language Inference Using Text Morphing Negru, Vlad Andrei Vacareanu, Robert Lemnaru, Camelia Surdeanu, Mihai Potolea, Rodica Computation and Language Artificial Intelligence We introduce MorphNLI, a modular step-by-step approach to natural language inference (NLI). When classifying the premise-hypothesis pairs into {entailment, contradiction, neutral}, we use a language model to generate the necessary edits to incrementally transform (i.e., morph) the premise into the hypothesis. Then, using an off-the-shelf NLI model we track how the entailment progresses with these atomic changes, aggregating these intermediate labels into a final output. We demonstrate the advantages of our proposed method particularly in realistic cross-domain settings, where our method always outperforms strong baselines with improvements up to 12.6% (relative). Further, our proposed approach is explainable as the atomic edits can be used to understand the overall NLI label. |
| title | MorphNLI: A Stepwise Approach to Natural Language Inference Using Text Morphing |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2502.09567 |