10 papers
UnifiedAttack: Evaluating the Safety of Large Multimodal Models in Synergistic Harmful Image-Text Generation
Bingjun Luo, Jialin Guo, Tony Wang +1
As Large Multimodal Models (LMMs) transition toward natively unified architectures, evaluating their safety in synergistic harmful image-text generation tasks becomes a critical ch…
Video-RSI: Recursive Self-Improvement of Video Understanding Agents via Harness Evolution
Bingjun Luo, Jialin Guo, Siqi Li
Video understanding agents acquire evidence through an executable harness that controls what they observe and how they use those observations. However, execution traces contain onl…
Temporal State Transport in Video Generation: Diagnosing and Correcting Spectral Imbalance
Luyao Tang, Bingjun Luo, Dong Yi +5
Reliable video generation requires more than high-quality frames to form a coherent story: a model must maintain a persistent state, transporting visual attributes such as identity…
SkillTrace: Traversing a Query-Skill Graph for Composable LLM Agents
Yue Yao, Shengyuan Wang, Xin Chen +4
Large language model agents increasingly solve complex tasks by composing reusable skills from a library. To address this, the key challenge is not merely to retrieve individually…
Adversarial Style Optimization: Enhancing VLM Jailbreaks by GRPO-based Stylistic Triggers Optimization
Bingjun Luo, Jialin Guo, Yue Yao +1
Multimodal Large Language Models (MLLMs) have achieved impressive performance, but their safety alignment remains vulnerable to jailbreak attacks. Existing content-based jailbreaks…
ST-SimDiff: Balancing Spatiotemporal Similarity and Difference for Efficient Video Understanding with MLLMs
Bingjun Luo, Tony Wang, Chaoqi Chen +1
Multimodal Large Language Models (MLLMs) face significant computational overhead when processing long videos due to the massive number of visual tokens required. To improve efficie…
Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding
Bingjun Luo, Tony Wang, Hanqi Chen +1
Recent advances in Multimodal Large Language Models (MLLMs) have significantly advanced video understanding tasks, yet challenges remain in efficiently compressing visual tokens wh…
HunyuanOCR Technical Report
Hunyuan Vision Team, Pengyuan Lyu, Xingyu Wan +29
This paper presents HunyuanOCR, a commercial-grade, open-source, and lightweight (1B parameters) Vision-Language Model (VLM) dedicated to OCR tasks. The architecture comprises a Na…
Unleashing the Intrinsic Visual Representation Capability of Multimodal Large Language Models
Hengzhuang Li, Xinsong Zhang, Qiming Peng +6
Multimodal Large Language Models (MLLMs) have demonstrated remarkable proficiency in multimodal tasks. Despite their impressive performance, MLLMs suffer from the modality imbalanc…
VITA-E: Natural Embodied Interaction with Concurrent Seeing, Hearing, Speaking, and Acting
Xiaoyu Liu, Chaoyou Fu, Chi Yan +15
Current Vision-Language-Action (VLA) models are often constrained by a rigid, static interaction paradigm, which lacks the ability to see, hear, speak, and act concurrently as well…