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 Shift Toward Open and Reproducible AI Research," Under Review, 2026. doi: 10.48550/arXiv.2606.16974..
UDH: Universal Deep Hiding for Steganography, Watermarking, and Light Field Messaging
Authors: Chaoning Zhang, Philipp Benz, Adil Karjauv, Geng Sun, In So Kweon
NeurIPS 2020 | Venue PDF | LLM Run Details | Input Tokens: 17,550 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,849 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We perform extensive analysis and demonstrate that the success of deep steganography can be attributed to a frequency discrepancy between C and the encoded secret image. Despite S being hidden in a cover-agnostic manner, strikingly, UDH achieves a performance comparable to the existing DDH. ... We co-train H and R on the Image Net [13] training dataset with the ADAM optimizer [31]. The APD (average pixel discrepancy) performance evaluated on the Image Net validation dataset is available in Table 1. |
| Researcher Affiliation | Academia | Chaoning Zhang KAIST EMAIL Philipp Benz KAIST EMAIL Adil Karjauv KAIST EMAIL Geng Sun KAIST EMAIL In So Kweon KAIST EMAIL |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Code: https://github.com/Chaoning Zhang/Universal-Deep-Hiding |
| Open Datasets | Yes | We co-train H and R on the Image Net [13] training dataset with the ADAM optimizer [31]. |
| Dataset Splits | Yes | The APD (average pixel discrepancy) performance evaluated on the Image Net validation dataset is available in Table 1. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running its experiments. It mentions using 'DNN-based' methods but no hardware specifications. |
| Software Dependencies | No | The paper mentions using the 'ADAM optimizer' and 'U-Net from Cycle-GAN' but does not specify their version numbers or any other software dependencies with version details. |
| Experiment Setup | Yes | The optimization goal is to minimize the loss de๏ฌned as L(S, Se, S ) = ||Se|| + ฮฒ||S S||, where Se = C C and following [2] we set ฮฒ to 0.75. ... The image resolution size is set to 128 128. |