5 papers
IV-CoT: Implicit Visual Chain-of-Thought for Structure-Aware Text-to-Image Generation
Zixuan Li, Haokun Lin, Yicheng Xiao +10
Unified multi-modal large language models (MLLMs) have achieved strong text-to-image generation quality, but still struggle with structure-aware prompt following, where object coun…
Attention Amnesia in Hybrid LLMs: When CoT Fine-Tuning Breaks Long-Range Recall, and How to Fix It
Xinyu Zhou, Boyu Zhu, Yi Xu +4
Chain-of-thought (CoT) supervised fine-tuning (SFT) is widely adopted to improve reasoning ability, yet we find that it systematically degrades long-context recall in hybrid linear…
AGC: Adaptive Geodesic Correction for Adversarial Robustness on Vision-Language Models
Zhiwei Li, Jiacheng Xue, Weining Wang +4
Vision-language models like CLIP have demonstrated remarkable zero-shot transfer capabilities. However, their susceptibility to imperceptible adversarial perturbations remains a cr…
TTP: Test-Time Padding for Adversarial Detection and Robust Adaptation on Vision-Language Models
Zhiwei Li, Yitian Pang, Weining Wang +2
Vision-Language Models (VLMs), such as CLIP, have achieved impressive zero-shot recognition performance but remain highly susceptible to adversarial perturbations, posing significa…
3SGen: Unified Subject, Style, and Structure-Driven Image Generation with Adaptive Task-specific Memory
Xinyang Song, Libin Wang, Weining Wang +6
Recent image generation approaches often address subject, style, and structure-driven conditioning in isolation, leading to feature entanglement and limited task transferability. I…