Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..

Boosting Generative Image Modeling via Joint Image-Feature Synthesis

Authors: Theodoros Kouzelis, Efstathios Karypidis, Ioannis Kakogeorgiou, Spyridon Gidaris, Nikos Komodakis

NeurIPS 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Evaluated in both conditional and unconditional settings, our method delivers substantial improvements in image quality and training convergence speed, establishing a new direction for representation-aware generative modeling. [...] 4 Experiments
Researcher Affiliation Collaboration Theodoros Kouzelis Archimedes, Athena RC National Technical University of Athens Efstathios Karypidis Archimedes, Athena RC National Technical University of Athens Ioannis Kakogeorgiou Archimedes, Athena RC IIT, NCSR "Demokritos" Spyros Gidaris valeo.ai Nikos Komodakis Archimedes, Athena RC University of Crete IACM-Forth
Pseudocode No The paper does not contain any section explicitly labeled "Pseudocode" or "Algorithm", nor are there any structured algorithm blocks presented as figures or within the main text.
Open Source Code Yes Project page and code: https://representationdiffusion.github.io/
Open Datasets Yes We follow the standard training setup of Di T (Peebles & Xie, 2023) and Si T (Ma et al., 2024), training on Image Net at 256 256 resolution with a batch size of 256.
Dataset Splits Yes We follow the standard training setup of Di T (Peebles & Xie, 2023) and Si T (Ma et al., 2024), training on Image Net at 256 256 resolution with a batch size of 256. [...] To benchmark generative performance, we report Frechet Inception Distance (FID) (Heusel et al., 2017), s FID (Nash et al., 2021), Inception Score (IS) (Salimans et al., 2016), Precision (Pre.) and Recall (Rec.) (Kynkäänniemi et al., 2019) using 50k samples and the ADM s Tensor Flow evaluation suite (Dhariwal & Nichol, 2021).
Hardware Specification Yes For both training and sampling we use 8 NVIDIA A100 40GB GPUs.
Software Dependencies No The paper mentions using "ADM s Tensor Flow evaluation suite" but does not specify version numbers for TensorFlow or any other software dependencies such as Python, PyTorch, or CUDA.
Experiment Setup Yes Table 8: Optimization details. The optimization hyperparameters for both Di T and Si T models. Batch Size 256 Optimizer Adam W LR 10 4 (β1, β2) (0.9, 0.999). [...] For Representation Guidance, we set pdrop = 0.2, the guidance scale to wr = 1.5 for B models and wr = 1.1 for XL models.