On Task Scope and Information Retention in Source Coding
cs.IT, cs.LG, eess.IV, math.IT
Submitted: 2026-09-29
Updated: 2026-09-29
Code: https://github.com/rezafuru/Task-Scope-and-Information-Retention
Sources
- Source-Side Sufficiency for the Information Bottleneck: Exact Reduction and Finite-Block Equivalence
- Variational image compression with a scale hyperprior
- CompressAI: a PyTorch library and evaluation platform for end-to-end compression research
- Model-Aware Rate-Distortion Limits for Task-Oriented Source Coding
- UniTAC: Universal Task-Aware Compression via Weighted Distortion Measures
- DT-UFC: Universal Large Model Feature Coding via Peaky-to-Balanced Distribution Transformation
- Deep Residual Learning for Image Recognition
- Task-Aware Answer Preservation under Audio Compression for Large Audio Language Models
- Indirect Rate-Distortion Function of a Binary i.i.d Source
- Joint Autoregressive and Hierarchical Priors for Learned Image Compression
- Bayes-Sufficient Representations in Supervised Learning
- Le Cam meets LeCun: Deficiency and Generic Feature Learning
Related papers
- Language Modeling is Monotone Compression
- Clipped Affine Policy: Low-Complexity Near-Optimal Online Power Control for Energy Harvesting Communications over Fading Channels
- Discrepancy for Random Linear Codes
- A New Approach to Code Smoothing Bounds
- Contextual Memory-Enhanced Source Coding for Low-SNR Communications
- Symmetry-Enforced Quadratic Approximate-Degradability Bounds for Noisy Landau-Streater Channels