SOURCE-LINKED INTELLIGENCE

T2LSC-Bench: Benchmarking Localized Semantic Control in Text-to-Image Generation

arXiv · AI, language, vision and robotics · article · Sep 2, 2026 · UTC

Recent text-to-image models have become increasingly capable of rendering explicit text, but reliable localized text control requires more than generating the correct string. In applications such as product labeling, signage, and interface design, target text should be rendered within a designated text-bearing region without altering the predefined subject identity or surrounding scene semantics. We refer to violations of this requirement as target-text-associated semantic leakage, in which target-text semantics are expressed through non-textual visual content beyond the designated anchor. Exi

Read original source ↗ Open in workspace

recordType

paper

region

Global