Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Safety-Potential Pruning for Enhancing Safety Prompts Against VLM Jailbreaking Without Retraining
Saved in:
| Published in: | Transactions of the Association for Computational Linguistics |
|---|---|
| Format: | Online Article RSS Article |
| Published: |
2026
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1868553987767140352 |
|---|---|
| collection | WordPress RSS FRELIP Feed Integration |
| container_title | Transactions of the Association for Computational Linguistics |
| description | |
| discipline_display | Linguistics and Philology |
| discipline_facet | Linguistics and Philology |
| format | Online Article RSS Article |
| genre | Journal Article |
| id | rss_article:93206 |
| institution | FRELIP |
| journal_source_facet | Transactions of the Association for Computational Linguistics |
| last_indexed | 2026-06-20T21:45:41.950Z |
| publishDate | 2026 |
| publishDateSort | 2026 |
| record_format | rss_article |
| spellingShingle | Safety-Potential Pruning for Enhancing Safety Prompts Against VLM Jailbreaking Without Retraining Linguistics and Philology General Linguistics and Philology |
| sub_discipline_display | General |
| sub_discipline_facet | General |
| subject_display | Linguistics and Philology General Linguistics and Philology |
| subject_facet | Linguistics and Philology General Linguistics and Philology |
| title | Safety-Potential Pruning for Enhancing Safety Prompts Against VLM Jailbreaking Without Retraining |
| title_alt | Poda de potencial de seguridad para mejorar las indicaciones de seguridad contra el jailbreaking de VLM sin reentrenamiento Élagage basé sur le potentiel de sécurité pour renforcer les consignes de sécurité contre le jailbreaking VLM sans réentraînement Poda de potencial de segurança para aprimorar prompts de segurança contra jailbreaking de VLM sem retreinamento |
| title_auth | Safety-Potential Pruning for Enhancing Safety Prompts Against VLM Jailbreaking Without Retraining |
| title_es_txt | Poda de potencial de seguridad para mejorar las indicaciones de seguridad contra el jailbreaking de VLM sin reentrenamiento |
| title_fr_txt | Élagage basé sur le potentiel de sécurité pour renforcer les consignes de sécurité contre le jailbreaking VLM sans réentraînement |
| title_full | Safety-Potential Pruning for Enhancing Safety Prompts Against VLM Jailbreaking Without Retraining |
| title_fullStr | Safety-Potential Pruning for Enhancing Safety Prompts Against VLM Jailbreaking Without Retraining |
| title_full_unstemmed | Safety-Potential Pruning for Enhancing Safety Prompts Against VLM Jailbreaking Without Retraining |
| title_pt_txt | Poda de potencial de segurança para aprimorar prompts de segurança contra jailbreaking de VLM sem retreinamento |
| title_short | Safety-Potential Pruning for Enhancing Safety Prompts Against VLM Jailbreaking Without Retraining |
| title_sort | safety-potential pruning for enhancing safety prompts against vlm jailbreaking without retraining |
| topic | Linguistics and Philology General Linguistics and Philology |
| url | https://direct.mit.edu/tacl/article/doi/10.1162/TACL.a.702/137206/Safety-Potential-Pruning-for-Enhancing-Safety |