12 papers
TriGeoBench: A Multimodal Mathematical Geometry Benchmark with Figures and TikZ
Jing Wang, Yitong Yang, Hao Dou +6
Multimodal Large Language Models show strong vision-language capabilities, yet their geometry-centric reasoning remains insufficiently understood. Existing mathematical benchmarks…
FIRE-VLA: Failure-Informed Self-Evolution for Vision-Language-Action Models in Autonomous Driving
Hao Dou
Reinforcement learning improves autonomous-driving vision-language-action (VLA) models by evaluating trajectories sampled from the current policy. Group relative policy optimizatio…
When Do Fewer Visual Tokens Accelerate Multimodal Inference? A Break-Even Study Across Decision Locations and Hardware
Hao Dou, Ruiwen Tian
Fewer visual tokens do not guarantee lower end-to-end latency. We evaluate break-even with a reproducible protocol that accounts for decision overhead, shared work, and the operato…
WorldScape Policy 2.0: Empowering Steerable World Action Modeling with Reasoning-Augmented Memory and In-Context Learning
Haisheng Su, Zongdai Liu, Xin Jin +13
World Action Models (WAMs) offer a promising paradigm for robotic manipulation by jointly modeling visual state transitions and robot actions. However, existing WAMs are constraine…
CIGPO: Contextual Information-Gain Policy Optimization for Multi-Turn Evidence-Reading LLM Agents
Hao Dou
Training multi-turn evidence-reading agents with outcome-only reinforcement learning is unstable because intermediate turns receive little direct credit. In HotpotQA experiments wi…
DreamX-World 1.0: A General-Purpose Interactive World Model
DreamX Team, Yancheng Bai, Rui Chen +20
DreamX-World 1.0 is a general-purpose interactive text/image-to-video world model for controllable long-horizon generation. It supports camera navigation, revisits to previously ob…
RISE: Reliable Improvement in Self-Evolving Vision-Language Models
Chaoran Xu, Yingmao Miao, Pengfei Zhang +3
Vision-language models (VLMs) have achieved strong multimodal reasoning capabilities, but further improving them still relies heavily on large-scale human-constructed supervision f…
Omni-WorldBench: Towards a Comprehensive Interaction-Centric Evaluation for World Models
Meiqi Wu, Zhixin Cai, Fufangchen Zhao +13
Video--based world models have emerged along two dominant paradigms: video generation and 3D reconstruction. However, existing evaluation benchmarks either focus narrowly on visual…
DiffStyle3D: Consistent 3D Gaussian Stylization via Attention Optimization
Yitong Yang, Xuexin Liu, Yinglin Wang +4
3D style transfer enables the creation of visually expressive 3D content, enriching the visual appearance of 3D scenes and objects. However, existing VGG- and CLIP-based methods st…
PositionIC: Unified Position and Identity Consistency for Image Customization
Junjie Hu, Tianyang Han, Kai Ma +6
Recent subject-driven image customization excels in fidelity, yet fine-grained instance-level spatial control remains an elusive challenge, hindering real-world applications. This…
InfinityDrive: Breaking Time Limits in Driving World Models
Xi Guo, Chenjing Ding, Haoxuan Dou +3
Autonomous driving systems struggle with complex scenarios due to limited access to diverse, extensive, and out-of-distribution driving data which are critical for safe navigation.…
Semi-supervised Text-based Person Search
Daming Gao, Yang Bai, Min Cao +3
Text-based person search (TBPS) aims to retrieve images of a specific person from a large image gallery based on a natural language description. Existing methods rely on massive an…