Don't Learn, Ground: A Case for Natural Language Inference with Visual Grounding
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917096994111488 |
|---|---|
| author | Ignatev, Daniil Santeer, Ayman Gatt, Albert Paperno, Denis |
| author_facet | Ignatev, Daniil Santeer, Ayman Gatt, Albert Paperno, Denis |
| contents | We propose a zero-shot method for Natural Language Inference (NLI) that leverages multimodal representations by grounding language in visual contexts. Our approach generates visual representations of premises using text-to-image models and performs inference by comparing these representations with textual hypotheses. We evaluate two inference techniques: cosine similarity and visual question answering. Our method achieves high accuracy without task-specific fine-tuning, demonstrating robustness against textual biases and surface heuristics. Additionally, we design a controlled adversarial dataset to validate the robustness of our approach. Our findings suggest that leveraging visual modality as a meaning representation provides a promising direction for robust natural language understanding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_17358 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Don't Learn, Ground: A Case for Natural Language Inference with Visual Grounding Ignatev, Daniil Santeer, Ayman Gatt, Albert Paperno, Denis Computation and Language We propose a zero-shot method for Natural Language Inference (NLI) that leverages multimodal representations by grounding language in visual contexts. Our approach generates visual representations of premises using text-to-image models and performs inference by comparing these representations with textual hypotheses. We evaluate two inference techniques: cosine similarity and visual question answering. Our method achieves high accuracy without task-specific fine-tuning, demonstrating robustness against textual biases and surface heuristics. Additionally, we design a controlled adversarial dataset to validate the robustness of our approach. Our findings suggest that leveraging visual modality as a meaning representation provides a promising direction for robust natural language understanding. |
| title | Don't Learn, Ground: A Case for Natural Language Inference with Visual Grounding |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2511.17358 |