MétaCan
Menu
Back to cohort
Record W4416707339 · doi:10.1109/lcomm.2025.3637721

Backdoor Attacks on Semantic Communications Systems With Joint Source and Channel Coding

2025· article· W4416707339 on OpenAlexaff
Long Bao Le

Bibliographic record

VenueIEEE Communications Letters · 2025
Typearticle
Language
FieldComputer Science
TopicAdversarial Robustness in Machine Learning
Canadian institutionsInstitut National de la Recherche ScientifiqueUniversité du Québec à Montréal
Fundersnot available
KeywordsBackdoorJoint (building)Focus (optics)Coding (social sciences)Channel (broadcasting)EncoderCommunications system

Abstract

fetched live from OpenAlex

This paper studies the backdoor attack on the task-oriented digital semantic communication system. We propose TrojanDSC, a two-stage backdoor learning strategy. Specifically, in the first stage we focus the attack on the critical latent dimensions using an optimized latent trigger, and in the second stage we fine-tune the semantic encoder and classification layer. We evaluate TrojanDSC across a range of channel conditions, trigger types, and the fine-tuning defense, showing that it achieves an attack success rate (ASR) over 94% for the poison rates of 10% or more with negligible accuracy degradation on clean inputs and remains effective even after the fine-tuning defense.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow), Science and technology studies, Scholarly communication, Open science
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.961
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0010.002
Science and technology studies0.0040.002
Scholarly communication0.0010.001
Open science0.0080.004
Research integrity0.0000.002
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.044
GPT teacher head0.294
Teacher spread0.250 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueIEEE Communications LettersSame topicAdversarial Robustness in Machine LearningFrench-language works237,207