8 papers
Enhanced Knowledge Distillation for Detection Transformer via Teacher Prediction Refinement
Yitong Xing, Yuhao Cheng, Yanping Li +1
Detection Transformers (DETRs) achieve strong performance in object detection but remain challenging to deploy on edge devices due to their high computational cost. Existing DETR d…
HarnessEval-W: Agentifying the Evaluation of Visual Worlds
Weiliang Chen, Haowen Sun, Jun Gao +40
A benchmark should deliver more than a scalar score: what makes an evaluation trustworthy is the reasoning that justifies the score. This is especially critical for world models, w…
Spatiotemporally Decoupled Autoregressive Diffusion Model for Human Motion Generation
Chengqun Yang, Liang Xu, Yanping Li +4
Text-driven human motion synthesis has made substantial development with two core modules of motion representation and generative architecture. For representation, Vector Quantizat…
Enfold: Folding World Model Imagination into Predictive Representations for Ultra-Efficient Embodied Control
Weili Zeng, Yitong Xing, Fulong Liu +10
World generative models are typically used through what they produce: a rendered future, a video-conditioned action, or latent context computed by a costly generative branch. We ar…
Posterior Continuation with Noise-Conditioned Frequency Exposure for Diffusion Inverse Problems
Feng Tian, Yixuan Li, Weili Zeng +3
Diffusion posterior sampling solves inverse problems by combining a pretrained diffusion prior with measurement-consistency guidance. However, full-band guidance can be unreliable…
SoulX-Podcast: Towards Realistic Long-form Podcasts with Dialectal and Paralinguistic Diversity
Hanke Xie, Haopeng Lin, Wenxiao Cao +17
Recent advances in text-to-speech (TTS) synthesis have significantly improved speech expressiveness and naturalness. However, most existing systems are tailored for single-speaker…
Flow Matching in the Low-Noise Regime: Pathologies and a Contrastive Remedy
Weili Zeng, Yichao Yan
Flow matching has recently emerged as a powerful alternative to diffusion models, providing a continuous-time formulation for generative modeling and representation learning. Yet,…
Skip-Vision: Efficient and Scalable Acceleration of Vision-Language Models via Adaptive Token Skipping
Weili Zeng, Ziyuan Huang, Kaixiang Ji +1
Transformer-based models have driven significant advancements in Multimodal Large Language Models (MLLMs), yet their computational costs surge drastically when scaling resolution,…