14 papers
Code2Worlds-2: Physically Plausible Agent Harness for Embodied World Simulation
Haoyu Zhang, Siming Li, Zeyu Zhang +3
Recent coding large language models (LLMs) can synthesize dynamic 4D worlds from natural language, but the resulting content remains a visual artifact. It lacks physical semantics…
Balancing Emotional Alignment and Semantic Consistency in Image Generation via Reinforcement Learning with Valence-Arousal Anchoring
Jisheng Dang, Zhenxuan Wang, Bin Li +3
Continuous emotion control in text-to-image generation requires a model to improve affective alignment without changing the objects, layout, or scene described by the prompt. Exist…
Multi-Agent Self-Improving Reinforcement Learning for Video Reasoning
Mingwen Zhang, Jisheng Dang, Minqiang Yang +3
Video reasoning tasks such as grounded video question answering and temporal grounding require selecting temporal evidence that supports the query. In many current training setups,…
On-Policy Self-Distillation in Diffusion Models
Wei Zhou, Xiongwei Zhu, Lingdong Kong +14
Reinforcement learning can align diffusion models with human preferences and task-specific objectives, but endpoint rewards do not specify how an intermediate denoising prediction…
A Robust Point Cloud Analysis Framework Inspired By Primary Visual Cortex
Jisheng Dang, Dengyue Pan, Delin Deng +6
Despite significant advancements in point cloud analysis, reducing energy consumption and improving robustness remain understudied, largely due to the inherent limitations of Convo…
WorldArena 2.0: Extending Embodied World Model Benchmarking on Modality, Functionality and Platform
Yu Shang, Yinzhou Tang, Yiding Ma +26
World models have emerged as a central paradigm for embodied intelligence, enabling agents to predict action-conditioned future and reason about environmental dynamics. However, ex…
AI for Auto-Research: Roadmap & User Guide
Lingdong Kong, Xian Sun, Wei Chow +17
AI-assisted research is crossing a threshold: fully automated systems can now generate research papers for as little as $15, while long-horizon agents can execute experiments, draf…
ReCALL: Recalibrating Capability Degradation for MLLM-based Composed Image Retrieval
Tianyu Yang, Chenwei He, Xiangzhao Hao +6
Composed Image Retrieval (CIR) aims to retrieve target images based on a hybrid query comprising a reference image and a modification text. Early dual-tower Vision-Language Models…
The Missing Half: Unveiling Training-time Implicit Safety Risks Beyond Deployment
Zhexin Zhang, Yida Lu, Junfeng Fang +8
Safety risks of AI models have been widely studied at deployment time, such as jailbreak attacks that elicit harmful outputs. In contrast, safety risks emerging during training rem…
Towards Comprehensive Stage-wise Benchmarking of Large Language Models in Fact-Checking
Hongzhan Lin, Zixin Chen, Zhiqi Shen +5
Large Language Models (LLMs) are increasingly deployed in real-world fact-checking systems, yet existing evaluations focus predominantly on claim verification and overlook the broa…
Thinking with Blueprints: Assisting Vision-Language Models in Spatial Reasoning via Structured Object Representation
Weijian Ma, Shizhao Sun, Tianyu Yu +3
Spatial reasoning -- the ability to perceive and reason about relationships in space -- advances vision-language models (VLMs) from visual perception toward spatial semantic unders…
EQ-TAA: Equivariant Traffic Accident Anticipation via Diffusion-Based Accident Video Synthesis
Jianwu Fang, Lei-Lei Li, Zhedong Zheng +4
Traffic Accident Anticipation (TAA) in traffic scenes is a challenging problem for achieving zero fatalities in the future. Current approaches typically treat TAA as a supervised l…
Understanding Long Videos via LLM-Powered Entity Relation Graphs
Meng Chu, Yicong Li, Tat-Seng Chua
The analysis of extended video content poses unique challenges in artificial intelligence, particularly when dealing with the complexity of tracking and understanding visual elemen…
3D-TAFS: A Training-free Framework for 3D Affordance Segmentation
Meng Chu, Xuan Zhang, Zhedong Zheng +1
Translating high-level linguistic instructions into precise robotic actions in the physical world remains challenging, particularly when considering the feasibility of interacting…