paper-with-me

Papers

AViTS: Adaptive Spatiotemporal Token Selection for Efficient Dynamic-Resolution Generation

2026-08-18 · Haoran Qin, Zhengan Yan, Shikang Zheng, Xiaobing Tu, Jiacheng Liu, Yuqi Lin, Chang Zou, JinShan Liu, Peiliang Cai, Xiantao Zhang, Jinkui Ren, Linfeng Zhang arxiv

Diffusion Transformers (DiTs) achieve high-quality generation but are costly due to iterative sampling. Dynamic-resolution sampling reduces early-stage cost by denoising at low resolution; however, uniformly upsampling all latent tokens at resolution transitions incurs redundant computation and may degrade fine-detail consistency. Existing partial upsampling strategies typically rely on local latent structure cues or single-step statistics, making it difficult to jointly capture token-text semantic relevance and token-wise representation dynamics across diffusion steps. We propose AViTS, an adaptive spatiotemporal token selection framework for dynamic-resolution DiTs. AViTS models spatial importance via latent-text attention and temporal importance via token-level feature variation across diffusion timesteps, and fuses them to enable spatiotemporal importance-aware selective upsampling: it prioritizes resolution refinement for critical tokens while deferring less important ones, thereby reducing redundant high-resolution computation and improving the quality-efficiency trade-off. AViTS achieves up to 6.34x on FLUX and nearly 9x FLOPs reduction on Qwen-Image-Edit and FLUX.1-Kontext-dev, orthogonal to distillation, quantization, and feature caching, and reaching 14.76x with distilled models. Code: https://github.com/QHR69/AViTS

📄 PDF Abstract BibTeX arXiv:2608.17995

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Stage-adaptive Token Selection for Efficient Omni-modal LLMs

2026-05-19 · Zijie Xin, Jie Yang, Ruixiang Zhao, Tianyi Wang 외 arxiv

Omni-modal large language models (om-LLMs) achieve unified audio-visual understanding by encoding video and audio into temporally aligned token sequences interleaved at the window level. However, processing these dense n…

PAVITS: Exploring Prosody-aware VITS for End-to-End Emotional Voice Conversion

2024-03-03 · Tianhua Qi, Wenming Zheng, Cheng Lu, Yuan Zong 외

In this paper, we propose Prosody-aware VITS (PAVITS) for emotional voice conversion (EVC), aiming to achieve two major objectives of EVC: high content naturalness and high emotional naturalness, which are crucial for me…

Voice Conversion

Going Down Memory Lane: Scaling Tokens for Video Stream Understanding with Dynamic KV-Cache Memory

2026-02-20 · Vatsal Agarwal, Saksham Suri, Matthew Gwilliam, Pulkit Kumar 외 arxiv

Streaming video understanding requires models to robustly encode, store, and retrieve information from a continuous video stream to support accurate video question answering (VQA). Existing state-of-the-art approaches re…

Video Question Answering

MATEY: multiscale adaptive foundation models for spatiotemporal physical systems

2024-12-29 · Pei Zhang, M. Paul Laiu, Matthew Norman, Doug Stefanski 외

Accurate representation of the multiscale features in spatiotemporal physical systems using vision transformer (ViT) architectures requires extremely long, computationally prohibitive token sequences. To address this iss…

Computational Efficiency

FlashVID: Efficient Video Large Language Models via Training-free Tree-based Spatiotemporal Token Merging

2026-02-08 · Ziyang Fan, Keyu Chen, Ruilong Xing, Yulin Li 외 arxiv

Although Video Large Language Models (VLLMs) have shown remarkable capabilities in video understanding, they are required to process high volumes of visual tokens, causing significant computational inefficiency. Existing…