AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

COSMIO: A Benchmark for Cross-Survey Modality Imputation

Dichang Zhang, Yixuan Shao, Jiali Cui, Haotian Yin, Yuanpeng Liu, Jiaiq Deng, Xiang Gao, Zhiqiang Lao, Heather Yu, Simon Birrer, Dimitris Samaras

NeurIPS 2026 · Vision-language and multimodal › Cross-modal learning · approach: Medical multimodal fusion · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Introduces COSMIO, a benchmark for cross-survey modality imputation over aligned multi-survey astronomical observations, where each object is seen by only a subset of surveys.

Problem

Large-scale astronomical surveys yield heterogeneous, partially observed multi-survey data: each object is observed by only some surveys, and surveys differ in resolution, wavelength coverage, noise, and footprint. Missing survey modalities must be imputed.

Approach

Formalizes the setting as cross-survey modality imputation and provides a benchmark of aligned multi-survey observations to evaluate imputation of missing survey modalities.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use MRI-specific imputation architectures instead of simple diffusionmedical imputation baselines (MMGAN, UMMI, DiffM4RI) · COSMIO common galaxies -> RomanCRPS: 0.159 (MMGAN) vs 0.069 (Asyn-Diff)Hurt
use asynchronous modality-level diffusion (Asyn-Diff) instead of pattern-specific Palettediffusion imputation · COSMIO common galaxies -> RomanCRPS: 0.072 -> 0.069Improved
train unified single model (RePaint) instead of pattern-specific Palettediffusion imputation · common galaxies -> RomanCRPS: 0.072 -> 0.090Hurt
evaluate on rare strong lenses instead of common galaxies (zero-shot)Asyn-Diff · Roman targetCRPS: 0.069 -> 0.090Hurt
evaluate on rare strong lenses instead of common galaxiesPalette · Roman targetCRPS: 0.072 -> 0.095Hurt
use imputed Roman images from better imputationdownstream lens classifier · Roman lens detectionAUPRC: 0.5027 (RePaint) -> 0.6104 (Asyn-Diff)Improved
lower assumed lens prevalence from 0.1 to 0.001Asyn-Diff classifier · Roman lens detectionAUPRC: 0.6104 -> 0.0411Hurt
physically motivated baselines instead of learned diffusionPSF-Linear · Euclid -> LSSTRMSE: 0.285 vs 0.332 (Asyn-Diff)Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Experimental setup

Benchmarks: COSMIO

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy