← 研踪 PaperTrailarXiv:2611.20381 · v1
Trait Collapse in Text-to-Image Generation: Evaluating and Mitigating Personality Inconsistency in Visual Storytelling
首次提交 2026-11-16 09:16
版本更新 2026-11-16 09:16
雷达观测 2026-11-17 17:12
摘要
In visual storytelling, a believable character must not only look consistent across scenes but also \textit{behave} in ways that reflect their personality. Current text-to-image (T2I) models preserve visual appearance effectively, yet consistently fail to maintain personality-relevant behavior---a failure we term \textit{trait collapse}. We show that conventional metrics (CLIP Score, DINO similarity) are blind to this failure, and that abstract personality descriptors and scene-level emotion labels do not produce consistent improvements over neutral prompts across models. To systematically diagnose trait collapse, we introduce \textbf{ViStoryBench-Personality}, a story-level benchmark combining an expert-validated Visual Cues Lexicon (Fleiss' ) with a six-dimensional metric suite covering personality alignment, cross-scene stability, and behavioral coherence. Evaluating seven T2I systems, we confirm a persistent semantic gap between abstract trait labels and visual evidence, and find that visual identity consistency does not imply personality consistency. To bridge this gap, we propose \textbf{Trait-Behavior-Visual (TBV)}, a training-free prompting framework that decomposes personality traits into renderable visual constraints. TBV improves Personality Alignment Score by up to , with larger gains in models with stronger language-conditioning capacity.
作者备注:EMNLP 2026 Main Conference Paper. 26 pages, 5 figures
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-16 09:16 · 最新观测版本
Trait Collapse in Text-to-Image Generation: Evaluating and Mitigating Personality Inconsistency in Visual Storytelling
In visual storytelling, a believable character must not only look consistent across scenes but also \textit{behave} in ways that reflect their personality. Current text-to-image (T2I) models preserve visual appearance effectively, yet consistently fail to maintain personality-relevant behavior---a failure we term \textit{trait collapse}. We show that conventional metrics (CLIP Score, DINO similarity) are blind to this failure, and that abstract personality descriptors and scene-level emotion labels do not produce consistent improvements over neutral prompts across models. To systematically diagnose trait collapse, we introduce \textbf{ViStoryBench-Personality}, a story-level benchmark combining an expert-validated Visual Cues Lexicon (Fleiss' ) with a six-dimensional metric suite covering personality alignment, cross-scene stability, and behavioral coherence. Evaluating seven T2I systems, we confirm a persistent semantic gap between abstract trait labels and visual evidence, and find that visual identity consistency does not imply personality consistency. To bridge this gap, we propose \textbf{Trait-Behavior-Visual (TBV)}, a training-free prompting framework that decomposes personality traits into renderable visual constraints. TBV improves Personality Alignment Score by up to , with larger gains in models with stronger language-conditioning capacity.
查看此版本解读 →查看此版本原文 ↗