TuringViT: Making SOTA Vision Transformers Accessible to All

• URL: http://arxiv.org/abs/2606.24253v2
• Date: Fri, 26 Jun 2026 09:37:54 GMT
• Title: TuringViT: Making SOTA Vision Transformers Accessible to All
• Abstract summary: TuringViT addresses challenges with three key designs: Turing Linear Attention, VISTA-Curation, and dynamic-resolution pretraining.<n>TuringViT outperforms leading open-source ViT baselines with only 10% of the data.<n>Our scaling-law analysis shows that TuringViT continues to improve predictably with curated data scale, far from saturation.
• Score: 40.44858252534576
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Modern VLMs and VLA systems commonly adopt off-the-shelf ViTs such as SigLIP2 as visual encoders, but diverse downstream requirements in latency, temporal modeling, and VLM integration often call for customized SOTA-level ViTs. Training such encoders remains beyond the reach of much of the community, as it requires massive image-text data, while standard softmax attention makes high-resolution or dynamic-resolution pretraining prohibitively costly and often forces low-resolution pretraining followed by post-hoc adaptation. TuringViT addresses these challenges with three key designs: Turing Linear Attention (TLA) for efficient sequence modeling, VISTA-Curation to construct supervision-rich image-video training data, and native dynamic-resolution pretraining that supports flexible inputs from the start and transfers seamlessly to downstream VLMs. As a result, TuringViT outperforms leading open-source ViT baselines with only 10% of the data, achieves stronger downstream VLM performance, and delivers substantially better latency scaling on high-resolution inputs. Our scaling-law analysis further shows that TuringViT continues to improve predictably with curated data scale, far from saturation. Its fast adaptation, hardware-friendly design, and efficient deployment have made it a unified visual foundation across XPeng's AI systems. More broadly, TuringViT provides a reproducible pipeline that dramatically lowers the cost for the community to train, customize, and deploy SOTA-level ViTs, moving toward making such Vision Transformers accessible to all.

Related papers

• UltraViT: Latency-Optimized On-device Vision Encoder for Large Vision-Language Models [47.9122865431321]
  We present UltraViT, a vision encoder for LVLMs, explicitly designed and optimized for on-device performance.<n>Specifically, by taking into account real on-device latencies, we design a pyramidal architecture that strategically integrates and adapts heterogeneous spatial mixers.<n>To pre-train UltraViT, we propose a novel two-stage generative pre-training strategy: cultivating rich spatial features via dense distillation, followed by direct generative supervision from a capacity-mixed frozen LLM.
  arXiv  Detail & Related papers  (2026-07-25T21:31:40Z)
• ViT-AdaLA: Adapting Vision Transformers with Linear Attention [71.36851471416034]
  Vision Transformers (ViTs) based vision foundation models (VFMs) have achieved remarkable performance across diverse vision tasks.<n>Existing linear attention approaches for ViTs are typically trained from scratch, requiring substantial computational resources.<n>We propose ViT-AdaLA, a novel framework for effectively adapting and transferring prior knowledge from VFMs to linear attention ViTs.
  arXiv  Detail & Related papers  (2026-03-17T02:15:48Z)
• ViTMAlis: Towards Latency-Critical Mobile Video Analytics with Vision Transformers [28.741078014867323]
  We introduce ViTMAlis, a device-to-edge offloading framework for vision transformers (ViTs)<n>ViTMAlis reduces end-to-end offloading latency while improving user-perceived rendering accuracy.<n>We implement a fully functional prototype of ViTMAlis on commodity mobile and edge devices.
  arXiv  Detail & Related papers  (2026-01-29T07:43:12Z)
• InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input Vision-Language Models [49.08289742711585]
  We propose InfiniteVL, a linear-complexity VLM architecture that synergizes sliding window attention (SWA) with Gated DeltaNet.<n>We show that InfiniteVL achieves over 3.6times inference speedup while maintaining constant latency and memory footprint.<n>In streaming video understanding scenarios, it sustains a stable 24 FPS real-time prefill speed while preserving long-term memory cache.
  arXiv  Detail & Related papers  (2025-12-09T17:18:32Z)
• LLaVA-UHD v3: Progressive Visual Compression for Efficient Native-Resolution Encoding in MLLMs [52.24096832965001]
  We present LLaVA-UHD v3, an MLLM centered upon our proposed Progressive Visual Compression (PVC) method.<n>The PVC method can be seamlessly integrated into standard Vision Transformer (ViT) to enable efficient native-resolution encoding.<n>Building upon ViT-UHD, LLaVA-UHD v3 also achieves competitive performance to Qwen2-VL, while further reducing TTFT by 1.9x.
  arXiv  Detail & Related papers  (2025-11-26T08:11:10Z)
• DyMU: Dynamic Merging and Virtual Unmerging for Efficient VLMs [124.52164183968145]
  We present DyMU, an efficient, training-free framework that reduces the computational burden of vision-language models (VLMs)<n>Our approach comprises two key components. First, Dynamic Token Merging (DToMe) reduces the number of visual token embeddings by merging similar tokens based on image complexity.<n>Second, Virtual Token Unmerging (VTU) simulates the expected token sequence for large language models (LLMs) by efficiently reconstructing the attention dynamics of a full sequence.
  arXiv  Detail & Related papers  (2025-04-23T18:38:18Z)
• ECViT: Efficient Convolutional Vision Transformer with Local-Attention and Multi-scale Stages [0.0]
  Vision Transformers (ViTs) have revolutionized computer vision by leveraging self-attention to model long-range dependencies.<n>We propose the Efficient Convolutional Vision Transformer (ECViT), a hybrid architecture that effectively combines the strengths of CNNs and Transformers.
  arXiv  Detail & Related papers  (2025-04-21T03:00:17Z)
• ViTALiTy: Unifying Low-rank and Sparse Approximation for Vision Transformer Acceleration with a Linear Taylor Attention [23.874485033096917]
  Vision Transformer (ViT) has emerged as a competitive alternative to convolutional neural networks for various computer vision applications. We propose a first-of-its-kind algorithm- hardware codesigned framework, dubbed ViTALiTy, for boosting the inference efficiency of ViTs. ViTALiTy unifies both low-rank and sparse components of the attention in ViTs.
  arXiv  Detail & Related papers  (2022-11-09T18:58:21Z)
• An Empirical Study of Training End-to-End Vision-and-Language Transformers [50.23532518166621]
  We present METER(textbfMultimodal textbfEnd-to-end textbfTransformtextbfER), through which we investigate how to design and pre-train a fully transformer-based VL model. Specifically, we dissect the model designs along multiple dimensions: vision encoders (e.g., CLIP-ViT, Swin transformer), text encoders (e.g., RoBERTa, DeBERTa), multimodal fusion (e.g., merged attention vs. co-
  arXiv  Detail & Related papers  (2021-11-03T17:55:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.