Skip to content
DraconiticResearch
Search Sources API
190,851 studies31 sourcesUpdated 23 min ago
PreprintarXiv licencearXiv

The Hidden Perception Constraint in Task-Aware Compression

  • Sahan Liyanaarachchi
  • Semih Akkoc
  • Sennur Ulukus
  • Aylin Yener
Open at sourcePDFDOICite

Abstract

With the recent advancements of neural compressors, explicitly incorporating perception constraints into the design of compression schemes has gained significant attention. Traditionally, these perception constraints ensure that the distribution of the reconstruction does not significantly deviate from the distribution of the source, thus attesting to the perceptual quality of the reconstruction. In this work, we uncover several perception constraints that are naturally present in task-aware compression. In particular, we consider a problem where the primary task is reconstruction and the secondary task is classification (i.e., a statistical test). We study this problem at varying levels of domain information available to us and discuss how to utilize the naturally emerging perception constraints to design rate-minimal compression schemes that also maximize the utility of our secondary task. We show that in this setting, if the decision boundaries of the classifier are ill-defined (mismatch) for our source distribution, then matching onto a target distribution enhances our classification accuracy.

Related

  1. UniTAC: Universal Task-Aware Compression via Weighted Distortion Measures
    Homa Esfahanizadeh, Matin Mortaheb, Adeel Mahmood +2
    arXivPreprintCC BY-NC-SA 4.0Full textGeneral mathematicsarXiv
  2. Sequential Lossy Compression With Causal Conditional Perception
    Photios A. Stavrou, Zixuan He
    arXivPreprintCC BY 4.0Full textGeneral mathematicsarXiv
  3. A Constraint-Aware Generative Framework for Synthetic Origin-Destination Demand in Logistics Networks
    Leian Chen
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  4. Sparse Identification for Automatic Large-Scale Screening: A Constraint-Aware Framework with Ultra Fast Decoding Algorithm
    Jianing Li, Li Chai, Yingcheng Lai
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  5. Box-iLQR: Constraint-Aware Feedback via Finite-Barrier Trajectory Optimization
    Abhijeet, Suman Chakravorty
    arXivPreprintCC0Full textOptimisation & controlarXiv

Cite

BibTeXRIS
BibTeX
@misc{liyanaarachchi2026hidden,
  title = {{The Hidden Perception Constraint in Task-Aware Compression}},
  author = {Sahan Liyanaarachchi and Semih Akkoc and Sennur Ulukus and Aylin Yener},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.35684},
  eprint = {2609.35684},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.35684},
}
RIS
TY  - UNPB
TI  - The Hidden Perception Constraint in Task-Aware Compression
AU  - Sahan Liyanaarachchi
AU  - Semih Akkoc
AU  - Sennur Ulukus
AU  - Aylin Yener
PY  - 2026
DA  - 2026/09/28
PB  - arXiv
DO  - 10.48550/arxiv.2609.35684
UR  - https://arxiv.org/abs/2609.35684
AB  - With the recent advancements of neural compressors, explicitly incorporating perception constraints into the design of compression schemes has gained significant attention. Traditionally, these perception constraints ensure that the distribution of the reconstruction does not significantly deviate from the distribution of the source, thus attesting to the perceptual quality of the reconstruction. In this work, we uncover several perception constraints that are naturally present in task-aware compression. In particular, we consider a problem where the primary task is reconstruction and the secondary task is classification (i.e., a statistical test). We study this problem at varying levels of domain information available to us and discuss how to utilize the naturally emerging perception constraints to design rate-minimal compression schemes that also maximize the utility of our secondary task. We show that in this setting, if the decision boundaries of the classifier are ill-defined (mismatch) for our source distribution, then matching onto a target distribution enhances our classification accuracy.
ER  -