10 papers
Dual-Frontier: When Can an Agent Trust Its World Model?
Huatai Zhu, Qiang Chen, Ziqian Kou +5
Learned world models are becoming essential to general-purpose agents: by predicting action consequences, they support planning and decision-making while reducing reliance on costl…
FireWorldBench: Benchmarking Complex Physical World Intelligence through Coupled-Field Fire Dynamics
Qiang Chen, Hao Guo, Huatai Zhu +7
Understanding the physical world requires more than object recognition, scene description, and short-term visual prediction, as real-world physical systems involve multiple continu…
Decision-Weighted Flow Matching for Contextual Stochastic Optimization
Jize Xie, Haomiao Wu, Qiang Chen +2
Conditional generative models are increasingly used as scenario generators for stochastic optimization, but standard training objectives emphasize uniform distributional fit rather…
VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Model
Wenhao Li, Xiu Su, Yichao Cao +5
Vision-Language-Action (VLA) models have demonstrated remarkable capabilities and generalization in embodied manipulation. However, their decision-making relies on a fast, instinct…
Detect by Yourself: Self-Designing Agentic Workflows for Few-Shot Graph Anomaly Detection
Tairan Huang, Qiang Chen, Yili Wang +4
Graph anomaly detection aims to identify anomaly nodes in attributed graphs and plays an important role in real-world applications. However, existing graph anomaly detection method…
Tools as Continuous Flow for Evolving Agentic Reasoning
Tairan Huang, Siyu Shang, Qiang Chen +2
Large Language Models (LLMs) have demonstrated remarkable capabilities in orchestrating tools for reasoning tasks. However, existing methods rely on a step-wise paradigm that lacks…
LEGATO: Good Identity Unlearning Is Continuous
Qiang Chen, Chun-Wun Cheng, Xiu Su +5
Machine unlearning has become a crucial role in enabling generative models trained on large datasets to remove sensitive, private, or copyright-protected data. However, existing ma…
Graph Unlearning Meets Influence-aware Negative Preference Optimization
Qiang Chen, Zhongze Wu, Ang He +6
Recent advancements in graph unlearning models have enhanced model utility by preserving the node representation essentially invariant, while using gradient ascent on the forget se…
Identify, Isolate, and Purge: Mitigating Hallucinations in LVLMs via Self-Evolving Distillation
Wenhao Li, Xiu Su, Jingyi Wu +5
Large Vision-Language Models (LVLMs) have demonstrated remarkable advancements in numerous areas such as multimedia. However, hallucination issues significantly limit their credibi…
Decoupled Video Generation with Chain of Training-free Diffusion Model Experts
Wenhao Li, Yichao Cao, Xiu Su +5
Video generation models hold substantial potential in areas such as filmmaking. However, current video diffusion models need high computational costs and produce suboptimal results…