HackerPapers

arXiv:1712.02950cs.CV

CycleGAN, a Master of Steganography

Casey Chu, Andrey Zhmoginov, Mark Sandler

Abstract

CycleGAN (Zhu et al. 2017) is one recent successful approach to learn a transformation between two image distributions. In a series of experiments, we demonstrate an intriguing property of the model: CycleGAN learns to "hide" information about a source image into the images it generates in a nearly imperceptible, high-frequency signal. This trick ensures that the generator can recover the original sample and thus satisfy the cyclic consistency requirement, while the generated image remains realistic. We connect this phenomenon with adversarial attacks by viewing CycleGAN's training procedure as training a generator of adversarial examples and demonstrate that the cyclic consistency loss causes CycleGAN to be especially vulnerable to adversarial attacks.

arXiv comments
NIPS 2017, workshop on Machine Deception
Journal reference
DOI
Subjects
cs.CVComputer Vision and Pattern Recognition cs.LGMachine Learning stat.MLMachine Learning (Statistics)

On Hacker News

Linked by 4 identifiable commenters in 4 comments across 4 threads, from Sep 2024 to Oct 2026. Found in 2 HN submissions from Jan 2019 to Jan 2019. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments4 threads

  1. Write Like It's 1866: LLMs Relearn Telegraphese 1 comment

    • https://arxiv.org/abs/1712.02950 (2017) …if you’re curious and missed that one like I did. Snack-sized paper with lots of satisfying visual examples.

      alwaView on HNParent context

  2. No right to relicense this project 1 comment

    • Somewhat annoyingly, there's been research that suggests that models can pass information to each other via (effectively) steganographic techniques - specific but apparently harmless choices of tokens, wordings, and so on; see https://arxiv.org/abs/1712.02950 and…

      nneonneoView on HNParent context

  3. How attention sinks keep language models stable 1 comment

    • > Researchers had observed similar patterns in BERT, where "a surprisingly large amount of attention focuses on the delimiter token [SEP] and periods," which they argued was used by the model as a sort of no-op. The same summer at Meta, researchers studying vision transformers…

      CalavarView on HNParent context

  4. Minifying HTML for GPT-4o: Remove all the HTML tags 1 comment

    • > You could even get GPT to generate it This isn't a good idea, if you want a fair test. See https://gwern.net/doc/reinforcement-learning/safe/2023-krako..., specifically https://arxiv.org/abs/1712.02950.

      wizzwizz4View on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.