AI RESEARCH
DVSM: Decoder-only View Synthesis Model Done Right
arXiv CS.CV
•
ArXi:2605.29891v1 Announce Type: new Recent Large View Synthesis Models (LVSMs) advocate an encoder-decoder architecture that separates reconstruction and rendering into distinct networks. We re-examine this design. Through controlled experiments, we show that a decoder-only architecture, which represents scenes implicitly as a KV-cache, outperforms encoder-decoder variants while using fewer parameters at identical rendering complexity.