Hugging Face Trending Papers July 21, 2026

SWITi: Quantifying and Reducing Tiling Artifacts with Sliding Window Inner Tiling

Read the original on Hugging Face Trending Papers →

SWITi is a test-time method for reducing artifacts in tiled predictions, particularly for neural networks that learn posterior distributions from which solutions are sampled at inference time. Tiled predictions are unavoidable for large image data, and artifacts arise whenever tiles are smaller than a network's receptive field and when tiles are independent posterior samples.

Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.

arXiv AI

Jul 22

SWITi: Quantifying and Reducing Tiling Artifacts with Sliding Window Inner Tiling

arXiv:2607. 18990v1 Announce Type: cross Abstract: SWITi is a test-time method for reducing artifacts in tiled predictions, particularly for neural networks that learn posterior distributions from which solutions are sampled at inference time.

By Federico Carrara, Aman Kukde, Melisande Croft, Joran Deschamps, Florian Jug

arXiv AI

1d ago

FrescoDiffusion: 4K Image-to-Video with Prior-Regularized Tiled Diffusion

arXiv:2603. 17555v2 Announce Type: replace-cross Abstract: Diffusion-based image-to-video (I2V) models are increasingly effective, yet they struggle to scale to ultra-high-resolution inputs (e.

By Hugo Caselles-Dupr\'e, Mathis Koroglu, Guillaume Jeanneret, Arnaud Dapogny, Matthieu Cord

diffusion

arXiv AI

Jul 24

Benchmarking Unlearning for Vision Transformers

arXiv:2602. 20114v2 Announce Type: replace-cross Abstract: Machine unlearning (MU) refers to the post-training capability to remove (the influence of) training examples that are incorrect, biased, or leak sensitive/private information.

By Kairan Zhao, Iurie Luca, Peter Triantafillou

llms diffusion computer-vision benchmarks

arXiv AI

Jul 14

Next-Dense-Stride Prediction for Multimodal Autoregressive Visual Modeling

arXiv:2607. 09892v1 Announce Type: cross Abstract: We introduce DenseAR, a new generative paradigm that reformulates autoregressive image generation as coarse-to-fine next-dense-stride prediction using a compact single-scale tokenizer.

By Chicago Y. Park, Jialin Mao, Xiaojian Xu, Taha Kass-Hout, Ulugbek S. Kamilov, Cao Xiao

diffusion computer-vision nlp multimodal

arXiv Machine Learning

Jun 8

LRMIL: Efficient Low-Resolution Multiple Instance Learning via High-Resolution Knowledge Distillation for Whole Slide Image Classification

arXiv:2606. 06864v1 Announce Type: cross Abstract: Multiple instance learning (MIL) has become a standard paradigm for whole slide image (WSI) analysis in digital pathology, as it enables slide-level prediction without dense annotations.

By Yonghan Shin, Won-Ki Jeong

rag computer-vision efficiency benchmarks

arXiv AI

Aug 6

CARVE: Cross-Slice Anisotropic Reallocation of Visual Evidence for Efficient 3D Medical Volume Understanding

arXiv:2608. 04515v1 Announce Type: cross Abstract: Slice-based MLLMs leverage mature 2D encoders by representing 3D volumes as sequences of 2D slices.

By Zhenyu Yi, Qiang Hu, Zhenhao Li, Jiaxuan Zhao, Yusong Sun, Lichi Zhang

llms benchmarks

Related stories

SWITi: Quantifying and Reducing Tiling Artifacts with Sliding Window Inner Tiling

FrescoDiffusion: 4K Image-to-Video with Prior-Regularized Tiled Diffusion

Benchmarking Unlearning for Vision Transformers

Next-Dense-Stride Prediction for Multimodal Autoregressive Visual Modeling

LRMIL: Efficient Low-Resolution Multiple Instance Learning via High-Resolution Knowledge Distillation for Whole Slide Image Classification

CARVE: Cross-Slice Anisotropic Reallocation of Visual Evidence for Efficient 3D Medical Volume Understanding