Papers with text-to-image systems
FairCoT: Enhancing Fairness in Text-to-Image Generation via Chain of Thought Reasoning with Multimodal Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | FairCoT enhances fairness in text-to-image generative models by integrating iterative reasoning . experimental evaluations demonstrate FairCot significantly enhances diversity without sacrificing image quality or semantic fidelity. |
| Approach: | FairCoT is a framework that enhances fairness in text-to-image generative models . it employs iterative CoT refinement to mitigate biases and dynamically adjusts textual prompts . |
| Outcome: | FairCoT combines iterative CoT refinement with iterating reasoning processes . it addresses limitations of zero-shot CoT in sensitive scenarios, authors say . |
CMIG: Conceptual Metaphor Theory-Inspired Framework for Metaphorical Image Generation (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing text-to-image systems often produce visually plausible but semantically literal outputs. |
| Approach: | They propose a structured prompting framework inspired by Conceptual Metaphor Theory . they propose to identify source–target mappings, filter projectable source attributes and select a visual realization strategy in a reproducible reasoning workflow. |
| Outcome: | The proposed framework improves semantic alignment and controllability on metaphor prompts. |