Padding Tone: A Mechanistic Analysis of Padding Tokens in T2I Models
Paper
•
2501.06751
•
Published
•
33
Michael Toker
tokeron
AI & ML interests
NLP, Computer Vision, Text to Image, LLMs, VLMs, Interpretability, Explainability
Recent Activity
upvoted
a
paper
about 1 month ago
RefVNLI: Towards Scalable Evaluation of Subject-driven Text-to-image
Generation
liked
a model
about 1 month ago
fal/AuraFlow
upvoted
a
paper
about 1 month ago
Multi-Domain Explainability of Preferences
Organizations
authored
a
paper
6 months ago
authored
4
papers
9 months ago
LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations
Paper
•
2410.02707
•
Published
•
49
A Dataset for Metaphor Detection in Early Medieval Hebrew Poetry
Paper
•
2402.17371
•
Published
Diffusion Lens: Interpreting Text Encoders in Text-to-Image Pipelines
Paper
•
2403.05846
•
Published
•
1
NL-Eye: Abductive NLI for Images
Paper
•
2410.02613
•
Published
•
23