# Towards Agnostic and Holistic Universal Image Segmentation with Bit Diffusion

Source: [Towards Agnostic and Holistic Universal Image Segmentation with Bit Diffusion](<https://www.sota2.com/research/paper/695de86853fdd5c361743a58>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-08T05:00:05.049Z

Authors: Jakob L{\\o}nborg Christensen, Morten Rieger Hannemose, Anders Bjorholm Dahl, Vedrana Andersen Dahl

Published: 2026-01-06T10:07:14.000Z

## Abstract

This paper introduces a diffusion-based framework for universal image segmentation, making agnostic segmentation possible without depending on mask-based frameworks and instead predicting the full segmentation in a holistic manner. We present several key adaptations to diffusion models, which are important in this discrete setting. Notably, we show that a location-aware palette with our 2D gray code ordering improves performance. Adding a final tanh activation function is crucial for discrete data. On optimizing diffusion parameters, the sigmoid loss weighting consistently outperforms alternatives, regardless of the prediction type used, and we settle on x-prediction. While our current model does not yet surpass leading mask-based architectures, it narrows the performance gap and introduces unique capabilities, such as principled ambiguity modeling, that these models lack. All models were trained from scratch, and we believe that combining our proposed improvements with large-scale pretraining or promptable conditioning could lead to competitive models.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Entity Segmentation on EntitySeg subset (roughly 4% of the 10% data) (val)](<https://www.sota2.com/research/sota/entity-segmentation-on-entityseg-subset-roughly-4-percent-of-the-10-percent-data-val.md>): Entity Segmentation; EntitySeg subset (roughly 4% of the 10% data) (val); ARI: 0.672.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2601.02881>)

[PDF](<https://arxiv.org/pdf/2601.02881.pdf>)
