research GeoSVG-RL Geometry-aware reinforcement learning for layout-constrained text-to-SVG diagram generation. OptiSQL Executable SQL generation from optical tokens. SemVink Visual global thinking for VLM semantic understanding of optical illusions. LogoText Semantic entanglement between logos and text in vision-language models. TCM LLM Name Bias Large language models over-rely on names to identify Chinese drug ingredients. Replace in Translation Boosting concept alignment in counterfactual text-to-image generation. code HRM-Idioma Hierarchical reasoning model pretrained with language data.
