cd /news/machine-learning/thread-efficient-decoding-for-neural… · home topics machine-learning article
[ARTICLE · art-116176] src=arxiv.org ↗ pub= topic=machine-learning verified=true sentiment=· neutral

Thread-Efficient Decoding for Neural Texture Compression

Researchers propose a shared decoder MLP architecture with gradual decoder freezing and texture clustering for neural texture compression, reducing GPU thread divergence by 25%-52% while preserving rendering quality. Evaluated on over 500 textures and real scenes, the method achieves up to 8.48x speedup on the Radeon RX 9070 XT GPU compared to non-shared baselines.

read1 min views1 publishedAug 31, 2026

arXiv:2608.27888v1 Announce Type: new Abstract: Neural texture compression (NTC) achieves higher compression ratios than BCn formats but suffers from GPU thread divergence, which significantly reduces runtime performance. In this work, we propose a shared decoder MLP architecture -- trained with a gradual decoder freezing schedule -- combined with texture clustering to reduce thread divergence by 25%-52% while preserving rendering quality. We evaluate our method on over 500 textures and multiple real rendering scenes, demonstrating up to 8.48x speedup on the Radeon RX 9070 XT GPU compared to non-shared baselines. Our key contributions include: (1) a unified shared decoder architecture that reduces divergence by grouping textures; (2) a training recipe with gradual decoder freezing that improves stability and reconstruction accuracy; (3) a semantic clustering strategy using CLIP embeddings that groups similar textures for effective decoder sharing; and (4) comprehensive performance and ablation studies validating our approach.

── more in #machine-learning 4 stories · sorted by recency
── more on @radeon rx 9070 xt 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/thread-efficient-dec…] indexed:0 read:1min 2026-08-31 ·