Pragma-VL: Towards a Pragmatic Arbitration of Safety and Helpfulness in MLLMs
Multimodal Large Language Models (MLLMs) pose critical safety challenges, as they are susceptible not only to adversarial attacks such as jailbreaking but also to inadvertently generating harmful content for benign users. While internal safety alignment via Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) is a primary mitigation strategy, current methods often face a safety-utility trade-off: they either refuse benign queries out of excessive caution or overlook latent risks in cross-modal interactions. To resolve this, we introduce Pragma-VL, an end-to-end alignment algorithm that enables MLLMs to pragmatically arbitrate between safety and helpfulness. First, we enhance visua...
| Authors | Ming Wen, Kun Yang, Xin Chen et al. |
|---|---|
| Primary category | cs.LG |
| Published | 28 February 2026 |
| Last updated at source | 28 February 2026 |
Connected in the knowledge graph
- fjh-0527/PragmaSafe-BeavertailsDataset · cited by
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.