6 papers
Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models
Guy Kaplan, Michael Toker, Yuval Reif +2
Text-to-image generation models suffer from alignment problems, where generated images fail to accurately capture the objects and relations in the text prompt. Prior work has focus…
DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models
Mor Ventura, Michael Toker, Or Patashnik +2
Text-to-Image (T2I) models have advanced rapidly, yet they remain vulnerable to semantic leakage, the unintended transfer of semantically related features between distinct entities…
LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations
Hadas Orgad, Michael Toker, Zorik Gekhman +4
Large language models (LLMs) often produce errors, including factual inaccuracies, biases, and reasoning failures, collectively referred to as "hallucinations". Recent studies have…
Padding Tone: A Mechanistic Analysis of Padding Tokens in T2I Models
Michael Toker, Ido Galil, Hadas Orgad +4
Text-to-image (T2I) diffusion models rely on encoded prompts to guide the image generation process. Typically, these prompts are extended to a fixed length by adding padding tokens…
Diffusion Lens: Interpreting Text Encoders in Text-to-Image Pipelines
Michael Toker, Hadas Orgad, Mor Ventura +2
Text-to-image diffusion models (T2I) use a latent representation of a text prompt to guide the image generation process. However, the process by which the encoder produces the text…
NL-Eye: Abductive NLI for Images
Mor Ventura, Michael Toker, Nitay Calderon +3
Will a Visual Language Model (VLM)-based bot warn us about slipping if it detects a wet floor? Recent VLMs have demonstrated impressive capabilities, yet their ability to infer out…