Article

Kinematics-Gated Post-Training Quantization: A Biomechanical Feedback Loop for Mixed-Precision Compression of a Video-Diffusion Transformer

2026-08-18

Abstract excerpt

Post-training quantization (PTQ) compresses large video-diffusion transformers without fine-tuning, but it is validated almost exclusively against frame-averaged reconstruction quality, which is not designed to express whether the generated human motion remains physically plausible: a video can be close to its full-precision reference frame by frame and still depict implausible motion. We study this gap for Dynami...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 91525c9f-0d82-5dda-af8c-a3e70452c6fd
DOI: 10.20944/preprints202608.1258.v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Kinematics-Gated Post-Training Quantization: A Biomechanical Feedback Loop for Mixed-Precision Compression of a Video-Diffusion TransformerDOI 10.20944/preprints202608.1258.v1Read

Towards Human-Centered and Efficient Video Synthesis: A Survey of Multimodal Diffusion ModelsSemantic proximity Read

Dream Your Pose: Robust Human Pose Generation via Uncertainty-Aware Structural Reward ModelingSemantic proximity Read

MiniCausal-T2V: Towards Ultra-Low Latency and Memory-Efficient Causal Video Generation on Edge DevicesSemantic proximity Read

A Latent Space Diffusion Transformer for High-Quality Video Frame InterpolationSemantic proximity Read

A Survey on Video Generation Technologies, Applications, and Ethical ConsiderationsSemantic proximity Read

Hybrid Diffusion Framework for Realistic Virtual Garment Try-OnSemantic proximity Read

Hierarchical Coarse-to-Fine cGAN for Subtype-Specific Freezing of Gait Signal GenerationSemantic proximity Read

A Resource-Efficient Approach to Text-Conditional Chest X-ray Generation Using Latent Diffusion ModelsSemantic proximity Read

Development of User-Customized Online Teaching Technology based on GPT, Enhanced by Generative AI-based Motion RecognitionSemantic proximity Read

Task-Adaptive Decoder Specialization for Unified Image Coding Toward Human Reconstruction and Machine ClassificationSemantic proximity Read

GaitEncoder: A Foundation Model of Gait Kinematics for Diverse Clinical Applications and PathologiesSemantic proximity Read

GestureFM : Compact Latent Flow Matching for Low-Latency Co-Speech Body-Motion GenerationSemantic proximity Read