독서리서치레이더투자 프레임워크
로그인 / 회원가입
← 읽기로 돌아가기
뉴스주제 · VisTA视觉弃权模型중대1 분 소요DisconfirmAI 편집부

Multimodal Models Learn to Refuse Impossible Edits

New research introduces visual abstention; the VisTA method enables models to refuse 93% of infeasible drawing requests while maintaining high editing accuracy.

2026-10-07 12:00 GMT+8

Unified multimodal models (UMMs) can now identify impossible drawing instructions and actively refuse them, rather than generating erroneous images.

Previously, even the strongest editing models refused only 0.4% of logically infeasible modification requests, often hallucinating non-existent objects or silently altering the prompt. Explicitly prompting models to report infeasibility increased refusals but reduced editing accuracy.

The VisTA training method proposed by Chufan Shi et al. pairs feasible and infeasible examples, forcing the model to judge feasibility before generating output. The resulting VisTA-BAGEL model refuses 93.0% of infeasible requests without reminders, while completing 74.3% of feasible edits—outperforming all eight evaluated mainstream models.

These results rely on the authors' self-created Draw-or-Decline benchmark and have not yet been independently reproduced.

출처

  • arxiv.org↗
주제

VisTA视觉弃权模型

南加州大学与加州大学圣地亚哥分校团队提出VisTA(Visual Transformation and Abstention)训练方法,旨在解决统一多模态模型(UMMs)在图像编辑中无法识别“不可行指令”的问题。该研究正式定义了“视觉弃权”…

2 단계1 건업데이트됨 2026-10-08 13:14

로그인 / 회원가입
독서리서치레이더투자 프레임워크

현재까지의 과정

  1. 2026-10-06论文《Visual Abstention in Unified Multimodal Models》提交至arXiv,提出DoD基准和VisTA方法。arxiv.org ↗
  2. 2026-10-08新闻发布指出VisTA使模型能识别并拒绝93%的不可行绘图请求,同时保持高编辑准确率。arxiv.org ↗

아직 해결되지 않은 문제

  • VisTA方法是否适用于其他类型的多模态生成任务(如视频编辑)?
  • 在实际部署中,如何平衡高拒绝率带来的用户体验影响与安全收益?
전체 주제 읽기 →
리서치 계속하기 →