CORTEXA
← Browse
arxivcs.CV2026-07-16

Introspective Attention Modulation for Safe Text-to-Image Generation

Basim Azam, Hossein Rahmani, Naveed Akhtar

State-of-the-art flow based text-to-image (T2I) models exhibit remarkable generative abilities but remain vulnerable to producing unsafe content. Prior safety efforts range from concept erasure and prompt filtering to classifier-based gating. However, simple techniques like parameter efficient adaptations of the models easily bypass such guardrails. We introduce a unique principled approach that achieves safety by regulating the model's attention dynamics through inference-time introspection, exhibiting intrinsic robustness. Our method analyzes and rebalances attention activations throughout image synthesis, steering generations away from unsafe concepts while preserving semantic alignment. This introspective control ensures safety of deployed models. Across standard and adversarial safety benchmarks, our approach achieves remarkable safety scores while maintaining or even improving alignment and perceptual quality. Our results reveal that attention-space regulation offers a considerably more promising path to safer diffusion transformer based image generation than the existing concept erasing mechanism.Our code can be accessed at https://basim-azam.github.io/iam/

View free PDFSource page

Related papers

arxivcs.CV2026-07-14

CoBind: Stage-Aware Compositional Binding for Training-Free Text-to-Image Generation

Kaijie Chen, Ethan Caldwell, Mira Vossen, Julian Hartwell, Serena Whitlock, Adrian Bellamy

Diffusion-based text-to-image models often fail on complex prompts involving multiple entities, attributes, and relations, producing object omissions, incorrect attribute assignments, or reversed spatial layouts. Existing training-free methods mainly strengthen token-level attent…

View free PDFSource page
arxivcs.CV2026-07-13

Read It Back: Pretrained MLLMs Are Zero-Shot Reward Models for Text-to-Image Generation

Runhui Huang, Qihui Zhang, Zhe Liu, Yu Gao, Jie Wu, Hengshuang Zhao

In this paper, we propose SpectraReward, a training-free reward function that turns pretrained MLLMs into off-the-shelf reward models for image-generation reinforcement learning. Instead of asking the MLLM to judge a generated image or answer decomposed verification questions, Sp…

View free PDFSource page
arxivcs.LGcs.AIcs.CV2026-07-16

Multi-Axis Max@K Reinforcement Learning for Representative Diversity in Text-to-Image Generation

Ku Onoda, Paavo Parmas, Hiroki Furuta, Soichiro Nishimori, Yuta Oshima, Shohei Taniguchi, et al.

Text-to-image (T2I) models can synthesize realistic, prompt-aligned images, yet samples generated for the same prompt often cover only a small subset of visually distinct modes. This limits the diversity of images, and for person-centric prompts, can reflect or amplify demographi…

View free PDFSource page
arxivcs.CV2026-07-18

UniNDM: A Unified Noise-driven Detection and Mitigation Framework Against Sexual Content in Text-to-Image Generation

Yao Huang, Yitong Sun, Huanran Chen, Ruochen Zhang, Shouwei Ruan, Ranjie Duan, et al.

Despite the impressive generative capabilities of text-to-image diffusion models, they remain vulnerable to implicit sexual prompts, where subtle cues disguised as benign terms or adversarial tokens unexpectedly generate the inappropriate content due to model biases or latent cor…

View free PDFSource page
arxivcs.CV2026-07-20

MixDiffusion: Mixing Diffusion-based Uni-condition Text-to-Image Generation Models for Multi-condition Image Synthesis

Pengcheng Wan, Liang Han, Lin Xu, Bowen Xiao, Liqiang Nie

Recent advances in text-to-image (T2I) generation have enabled controllable image synthesis by incorporating conditions beyond text. However, most existing diffusion-based methods are limited to a single type of control condition (e.g., bounding boxes or keypoints), which restric…

View free PDFSource page
arxivcs.CVcs.GR2026-07-13

Latent-Identity Tuning in Text-to-Image Personalization Models

Daniel Garibi, Ronen Kamenetsky, Hadar Averbuch-Elor, Daniel Cohen-Or, Or Patashnik

Generating and editing a person's face demands high precision, as even minor modifications can significantly alter a subject's perceived identity. Current personalization and editing methods built on general-purpose text-to-image models, however, often lack the precision required…

View free PDFSource page