# S23DR 2026 Winning Solution

arXiv: https://arxiv.org/abs/2606.06695 · HTML: https://mlsanity.com/paper/2606.06695/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 4 of its 7 references that are in the map.
- First posted on arXiv 4 Jun 2026.

## Builds on (most cited first, 4 of 4)

- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [Explaining Human Preferences via Metrics for Structured 3D Reconstruction](https://mlsanity.com/paper/2503.08208/) (2025)

## Abstract (arXiv)

This text presents the winning solution to the S23DR 2026 challenge for structured 3D wireframe reconstruction from sparse SfM, fitted depth, and semantic segmentations. The method treats vertices as a conditional set and denoises 64 vertex tokens with a flow-matching DiT conditioned on Perceiver-style scene tokens. A global pass predicts the coarse structure, a hull-cropped second pass refines it, and a small multi-sample consensus step keeps the stochastic sampler well behaved. The final system ranked first on the private leaderboard, achievingHSS = 0.654.