Research map / Architectures and efficiency / Pruning and efficient tuning
Infomax Neural Joint Source-Channel Coding via Adversarial Bit Flip
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add adversarial bit flip to NECST | joint source-channel coding · BinaryMNIST, Omniglot, CIFAR10 and MNIST at ε=0.1–0.4 | squared reconstruction error per image | Helped |
| Add mutual information maximization to ABF (ablation) | 100-bit MNIST coding · MNIST, ε=0.1 | squared reconstruction error per image: 13.369 -> 13.251 | Helped |
| Add mutual information maximization to ABF (ablation) | 100-bit MNIST coding · MNIST, ε=0.2 | squared reconstruction error per image: 22.013 -> 22.026 | Hurt |
| Add mutual information maximization to ABF (ablation) | 100-bit MNIST coding · MNIST, ε=0.3 | squared reconstruction error per image: 33.319 -> 33.176 | Helped |
| Add mutual information maximization to ABF (ablation) | 100-bit MNIST coding · MNIST, ε=0.4 | squared reconstruction error per image: 48.041 -> 46.555 | Helped |
| Use IABF instead of NECST | joint source-channel coding · CIFAR10, ε=0.3 | squared reconstruction error per image: 89.619 -> 83.193 | Helped |
Closest papers by meaning
- Neural Joint Source-Channel Coding (ICML 2019 · similarity 0.89)
- SoftBinary Coding: A New Information-Theoretic Paradigm for Neural Compression via Fast Channel Simulation (ICML 2026 · similarity 0.84)
- Discrimination-aware Channel Pruning for Deep Neural Networks (NeurIPS 2018 · similarity 0.82)
- Approaching Rate-Distortion Limits in Neural Compression with Lattice Transform Coding (ICLR 2025 · similarity 0.82)
- Towards Lossless Implicit Neural Representation via Bit Plane Decomposition (CVPR 2025 · similarity 0.82)
- Cauchy-Schwarz Divergence Information Bottleneck for Regression (ICLR 2024 · similarity 0.81)
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.