technique

feed-forward transformer

techniqueactiveprovisionalfeed-forward-transformer-806122d9·1 events·first seen 16d ago

Aliases: feed-forward transformer

Co-occurring entities

power-law scaling CompVis Self-Supervised Learning RayDer novel view synthesis

More like this (12)

FeedForward submodule causal transformer transformer architecture autoregressive transformer Feedforward Neural Networks Differential Transformer Variable-Width Transformers flow-matching diffusion transformer Training-Free Looped Transformers Linear Diffusion Transformer Conditional Diffusion Transformer Universal Physics Transformer (UPT)

Recent events (1)

7arXiv · cs.LG·16d ago·source ↗

RayDer: Scalable Self-Supervised Novel View Synthesis via Unified Feed-Forward Transformer

RayDer is a unified feed-forward transformer that consolidates camera estimation, scene reconstruction, and rendering into a single backbone for self-supervised novel view synthesis (NVS). By treating dynamic content as a nuisance factor absorbed by a minimal dynamic state, it enables stable training on unconstrained real-world video without requiring dynamic-scene reconstruction. The model exhibits clean power-law scaling with both data and compute across multiple model sizes, and achieves zero-shot open-set performance competitive with supervised state-of-the-art methods on multiple benchmarks.

Training Infrastructure Frontier Model Releases feed-forward transformer power-law scaling CompVis +4 more