Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• ClipTTT adapts Large Vision-Language Models (LVLMs) under degraded conditions on the fly using a single test sample.method 95%
• ClipTTT leverages the image-text alignment strength of a pre-trained CLIP model as a stable guidance signal to identify reliable self-supervision targets.method 95%
• ClipTTT effectively mitigates hallucinations and improves descriptive faithfulness under visual corruptions.result 95%
• Visual corruptions act as additional distribution shifts, significantly amplifying hallucination rates in real-world applications.technical 90%
• ClipTTT was evaluated on standard hallucination benchmarks with 15 common corruptions.method 95%
• The ClipTTT framework uses a student-teacher model for on-the-fly adaptation, where the teacher generates caption candidates and CLIP scores them to select a pseudo-label for the student.method 95%
• Sentence-level averaging of CLIP scores is used to avoid truncation issues with CLIP's text encoder and ensure stable similarity estimates for longer captions.technical 90%
• ClipTTT adapts Large Vision-Language Models (LVLMs) under degraded conditions on the fly using a single test sample.method 95%
• ClipTTT leverages the image-text alignment strength of a pre-trained CLIP model as a stable guidance signal to identify reliable self-supervision targets.technical 95%
• ClipTTT effectively mitigates hallucinations and improves descriptive faithfulness under visual corruptions.result 95%
• Visual corruptions act as additional distribution shifts, significantly amplifying hallucination rates in real-world applications.technical 90%
• ClipTTT enables rapid adaptation without altering the base LVLMs.method 90%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card