About this paper
By baking global scene information from cross-view attention into a lightweight MLP, Test-Time Baking (TTB) enhances the fidelity of novel-view decoding in Large View Synthesis Models while maintaining the efficiency of Test-Time Training.
Decoder-only Large View Synthesis Models (LVSMs) with KV-cache have recently achieved state-of-the-art quality by regressing novel views from neural networks without reconstructing 3D geometry, while Test-Time Training (TTT) layers save computation but sacrifice quality. We propose Test-Time Baking (TTB), a strategy that learns to bake global scene information from cross-view attention into a lightweight MLP, enabling fast and high-fidelity novel-view decoding. The first key is to repurpose the fast-weight MLP of TTT to learn the cross-view attention mapping explicitly from its input-output pairs, rather than replacing attention entirely, achieving the same rendering efficiency as TTT…
Decoder-only Large View Synthesis Models (LVSMs)KV-cacheNovel viewsTest-Time Training (TTT)Test-Time Baking (TTB)Cross-view attentionLightweight MLPContext prefillingModality misalignmentContext tokens
- Upvotes
- 17
- Authors
- 9
- Published
- Nov 12, 2026
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.
Trending history
TTB: Test-time Baking for Efficient Rendering of Decoder-only View Synthesis Models was first seen on Hugging Face · Daily papers on Nov 13, 2026. Since then it has appeared in 35 checks on 4 different days. Its best position was #11 of 30, reached on Nov 16, 03:28 UTC. It is #11 right now, climbing from #12 to #11 over the last 6 checks.