Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2604.26419 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914516277657600 |
|---|---|
| author | Song, Junru Hu, Yimeng Chen, Yijing Li, Huining Li, Qian Cui, Lizhen Du, Yuntao |
| author_facet | Song, Junru Hu, Yimeng Chen, Yijing Li, Huining Li, Qian Cui, Lizhen Du, Yuntao |
| contents | Large Vision-Language Models (VLMs) have achieved remarkable multimodal performance yet remain prone to factual hallucinations, particularly in long-tail or specialized domains. Moreover, current models exhibit a weak capacity to refuse queries that exceed their parametric knowledge. In this paper, we propose a systematic framework to enhance the refusal capability of VLMs when facing such unknown questions. We first curate a model-specific "Visual-Idk" (Visual-I don't know) dataset, leveraging multi-sample consistency probing to distinguish between known and unknown facts. We then align the model using supervised fine-tuning followed by preference-aware optimization (e.g., DPO, ORPO) to effectively delineate its knowledge boundaries. Results on the Visual-Idk dataset show our method improves the Truthful Rate from 57.9\% to 67.3\%. Additionally, internal probing also demonstrates that the model genuinely recognizes its boundaries instead of just memorizing refusal patterns. Our framework further generalizes to out-of-distribution medical and perceptual domains, providing a robust path toward more trustworthy and prudent visual assistants. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_26419 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Delineating Knowledge Boundaries for Honest Large Vision-Language Models Song, Junru Hu, Yimeng Chen, Yijing Li, Huining Li, Qian Cui, Lizhen Du, Yuntao Computer Vision and Pattern Recognition Artificial Intelligence Large Vision-Language Models (VLMs) have achieved remarkable multimodal performance yet remain prone to factual hallucinations, particularly in long-tail or specialized domains. Moreover, current models exhibit a weak capacity to refuse queries that exceed their parametric knowledge. In this paper, we propose a systematic framework to enhance the refusal capability of VLMs when facing such unknown questions. We first curate a model-specific "Visual-Idk" (Visual-I don't know) dataset, leveraging multi-sample consistency probing to distinguish between known and unknown facts. We then align the model using supervised fine-tuning followed by preference-aware optimization (e.g., DPO, ORPO) to effectively delineate its knowledge boundaries. Results on the Visual-Idk dataset show our method improves the Truthful Rate from 57.9\% to 67.3\%. Additionally, internal probing also demonstrates that the model genuinely recognizes its boundaries instead of just memorizing refusal patterns. Our framework further generalizes to out-of-distribution medical and perceptual domains, providing a robust path toward more trustworthy and prudent visual assistants. |
| title | Delineating Knowledge Boundaries for Honest Large Vision-Language Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2604.26419 |