How the MPCT Multiscale Point Cloud Transformer Boosts 3D Classification Accuracy

This article reviews the MPCT framework—a multiscale point‑cloud transformer built on a residual network that leverages permutation‑invariant self‑attention, point‑enhancement, and hierarchical feature aggregation to achieve state‑of‑the‑art results on ModelNet40 and ScanObjectNN datasets.

3D classificationSelf-Attentionmultiscale

0 likes · 14 min read

How the MPCT Multiscale Point Cloud Transformer Boosts 3D Classification Accuracy

AIWalker

Mar 10, 2025 · Artificial Intelligence

FlexVAR: Autoregressive Image Generation with Inpainting and Speed‑Quality Control

FlexVAR replaces residual prediction with direct ground‑truth prediction in visual autoregressive modeling, enabling generation of arbitrary resolutions and aspect ratios, supporting image‑to‑image tasks such as inpainting and upscaling, and offering adjustable inference steps that trade speed for quality while achieving state‑of‑the‑art FID scores.

Autoregressive ModelsVQVAEflexvar

0 likes · 17 min read

FlexVAR: Autoregressive Image Generation with Inpainting and Speed‑Quality Control

Sohu Tech Products

Jun 7, 2023 · Artificial Intelligence

Multiscale PU Learning for Detecting AI‑Generated Text

Researchers from Peking University and Huawei present a multiscale positive‑unlabeled learning framework that significantly improves detection of AI‑generated short and long texts, addressing the difficulty of distinguishing AI‑written content from human writing and outperforming existing baselines on multiple benchmarks.

AI detectionLarge Language ModelsPu-Learning

0 likes · 8 min read

Multiscale PU Learning for Detecting AI‑Generated Text