11 papers
Rethinking Event-Based Object Detection through Representation-Level Temporal Aggregation and Model-Level Hypergraph Reasoning
Meisen Wang, Hao Deng, Wei Bao +6
Event cameras provide microsecond-level temporal resolution, low latency, and high dynamic range, offering potential for perception under fast motion and challenging illumination c…
When Policy Entropy Constraint Fails: Preserving Diversity in Flow-based RLHF via Perceptual Entropy
Xiaofeng Tan, Jun Liu, Bin-Bin Gao +5
RLHF is widely used to align flow-matching text-to-image models with human preferences, but often leads to severe diversity collapse after fine-tuning. In RL, diversity is often as…
L2P: Unlocking Latent Potential for Pixel Generation
Zhennan Chen, Junwei Zhu, Xu Chen +7
Pixel diffusion models have recently regained attention for visual generation. However, training advanced pixel-space models from scratch demands prohibitive computational and data…
Large-Scale Universal Defect Generation: Foundation Models and Datasets
Yuanting Fan, Jun Liu, Bin-Bin Gao +5
Existing defect/anomaly generation methods often rely on few-shot learning, which overfits to specific defect categories due to the lack of large-scale paired defect editing data.…
PET-DINO: Unifying Visual Cues into Grounding DINO with Prompt-Enriched Training
Weifu Fu, Jinyang Li, Bin-Bin Gao +6
Open-Set Object Detection (OSOD) enables recognition of novel categories beyond fixed classes but faces challenges in aligning text representations with complex visual concepts and…
ConsistentRFT: Reducing Visual Hallucinations in Flow-based Reinforcement Fine-Tuning
Xiaofeng Tan, Jun Liu, Yuanting Fan +7
Reinforcement Fine-Tuning (RFT) on flow-based models is crucial for preference alignment. However, they often introduce visual hallucinations like over-optimized details and semant…