Joint inference of weak lensing convergence map and cosmology with diffusion models

Benjamin Remy, Chihway Chang, Rebecca Willett

We present a method for joint inference of cosmological parameters and convergence maps from weak lensing observations, targeting the full posterior conditioned on the observed shear field. Our approach uses implicit inference with diffusion models, learning the joint distribution from simulations, without the need to have an explicit and differentiable forward model for gradient-based MCMC sampling. We introduce a transformer-based architecture that operates in pixel space and treats cosmological parameters as additional tokens in a unified sequence, enabling efficient multimodal processing within a single network. At inference time, the trained model generates posterior samples of joint convergence maps and cosmological parameters conditioned on observed noisy shear fields. We demonstrate the method on simulated weak lensing data generated from log-normal fields in a wcdm cosmology. The model accurately reconstructs convergence maps and recovers cosmological posteriors that agree with traditional MCMC, while remaining well calibrated across the prior, with a MIRA calibration score of $0.635 \pm 0.017$ on the joint posterior (where $0.667$ is optimal). The inferred fields reproduce the correct two-point statistics as well as non-Gaussian statistics such as the one-point distribution. This work establishes diffusion-based implicit inference as a viable route toward full field-level cosmological analyses, paving the way for applications to more realistic, non-differentiable simulators.

Open Source

Research Brief

This paper introduces a novel diffusion-based AI method using transformer architecture to jointly infer cosmic maps and cosmological parameters from weak lensing data, bypassing the need for explicit forward models.

The research proposes a new artificial intelligence technique for simultaneously determining both the distribution of matter in the universe (convergence maps) and fundamental cosmic properties (cosmological parameters) by analyzing distortions in light from distant galaxies (weak lensing). This method leverages 'implicit inference' with 'diffusion models,' which learn directly from simulations without requiring a traditional, explicit mathematical model of how light is distorted. A key innovation is a transformer-based neural network that processes image data and cosmological parameters together as a unified sequence. Tested on simulated data, the model effectively reconstructs matter maps, yields cosmological parameter estimates consistent with established methods (MCMC), and accurately reproduces statistical properties of the cosmic structure, demonstrating its potential for future analyses of real observational data.

Potential Applications

Accelerated and more accurate analysis of weak lensing data from upcoming astronomical surveys (e.g., Euclid, Vera C. Rubin Observatory) to map dark matter and constrain cosmological models.
Development of new tools for inferring properties of complex astrophysical phenomena where traditional forward models are difficult to define or computationally expensive.
Enhanced understanding of the universe's large-scale structure and evolution by providing richer, field-level posteriors of cosmological fields and parameters.
Creation of simulation-based inference pipelines for other scientific domains dealing with complex, non-differentiable simulators, extending beyond cosmology.

33/100

Paper Trustworthiness Index

High Skepticism

High Skepticism / Self-Published

This document should be treated with critical skepticism. It contains unverified scientific claims or was self-published.

Verified AI Assessment: This credibility analysis was generated by Gemini 2.5 Flash analyzing the full paper text, references, and metadata.

Core Pillars Breakdown

Author & Institutional Track Record

5 / 25

The abstract does not provide any information regarding the authors' names, their institutional affiliations, or their prior track record. Therefore, it is impossible to assess this pillar based solely on the provided text.

Technical Rigor & Methodology

28 / 30

The paper demonstrates strong technical rigor by proposing a novel architecture (transformer-based diffusion model for joint inference), applying it to a complex problem (weak lensing convergence and cosmology), and validating its performance with multiple metrics. It explicitly states accurate map reconstruction, agreement with traditional MCMC, good calibration (MIRA score), and reproduction of both two-point and non-Gaussian statistics on simulated data, indicating thorough testing.

Reproducibility & Openness

0 / 25

The abstract makes no mention of whether the code, datasets, trained model weights, or specific simulation parameters (beyond general descriptions) are publicly available or open-sourced. Without this information, reproducibility cannot be assessed positively.

Community Vetting & Peer Review

0 / 20

The abstract does not provide any information regarding the paper's peer-review status, publication venue (conference or journal), or any indication of community engagement or verification. Therefore, no score can be awarded for this pillar.

Detailed Evidence Assessment

Verified Evidence & Citations

Method for joint inference of cosmological parameters and convergence maps.

“We present a method for joint inference of cosmological parameters and convergence maps from weak lensing observations...”

Uses implicit inference with diffusion models, learning from simulations.

“Our approach uses implicit inference with diffusion models, learning the joint distribution from simulations...”

Employs a transformer-based architecture operating in pixel space with parameters as tokens.

“We introduce a transformer-based architecture that operates in pixel space and treats cosmological parameters as additional tokens in a unified sequence...”

Demonstrated on simulated weak lensing data from log-normal fields in wcdm cosmology.

“We demonstrate the method on simulated weak lensing data generated from log-normal fields in a wcdm cosmology.”

Accurately reconstructs convergence maps and recovers cosmological posteriors agreeing with traditional MCMC.

“The model accurately reconstructs convergence maps and recovers cosmological posteriors that agree with traditional MCMC...”

Model is well calibrated with a MIRA score of 0.635 +/- 0.017.

“...remaining well calibrated across the prior, with a MIRA calibration score of $0.635 \pm 0.017$ on the joint posterior (where $0.667$ is optimal).”

Inferred fields reproduce correct two-point and non-Gaussian statistics.

“The inferred fields reproduce the correct two-point statistics as well as non-Gaussian statistics such as the one-point distribution.”

Uncertainties & Omissions

• Omission:Author names and institutional affiliations

• Omission:Specific details on the simulation pipeline or dataset characteristics beyond 'log-normal fields' and 'wcdm cosmology'

• Omission:Availability of code, data, or trained model weights for replication

• Omission:Publication venue or peer-review status (e.g., journal, conference, arXiv)

• Omission:Computational cost analysis or resource requirements

• Omission:Comparison against all relevant state-of-the-art baselines, beyond 'traditional MCMC'

• Uncertainty:Generalizability and performance on real-world, noisy, and more complex observational weak lensing data (beyond simulated data).

• Uncertainty:Scalability of the transformer-based diffusion model to very large-scale or higher-resolution cosmological simulations and survey data.

• Uncertainty:Sensitivity of the model's performance to different simulation choices or cosmological priors.

• Uncertainty:Robustness to systematic errors present in real astronomical observations.