Pith. sign in

REVIEW 3 cited by

Backdoor Attacks on Pre-trained Models by Layerwise Weight Poisoning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2108.13888 v1 pith:IJN4WNZ7 submitted 2021-08-31 cs.CR cs.CL

Backdoor Attacks on Pre-trained Models by Layerwise Weight Poisoning

classification cs.CR cs.CL
keywords textbfmethodpoisoningtriggersappliedattacksbackdoorbackdoors
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

\textbf{P}re-\textbf{T}rained \textbf{M}odel\textbf{s} have been widely applied and recently proved vulnerable under backdoor attacks: the released pre-trained weights can be maliciously poisoned with certain triggers. When the triggers are activated, even the fine-tuned model will predict pre-defined labels, causing a security threat. These backdoors generated by the poisoning methods can be erased by changing hyper-parameters during fine-tuning or detected by finding the triggers. In this paper, we propose a stronger weight-poisoning attack method that introduces a layerwise weight poisoning strategy to plant deeper backdoors; we also introduce a combinatorial trigger that cannot be easily detected. The experiments on text classification tasks show that previous defense methods cannot resist our weight-poisoning method, which indicates that our method can be widely applied and may provide hints for future model robustness studies.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Lethe: Purifying Backdoored Large Language Models with Knowledge Dilution

    cs.CL 2025-08 conditional novelty 6.0

    LETHE uses parameter-level model merging plus prompt-level word definitions to dilute backdoor behavior in LLMs, cutting attack success to below 7% in most tested settings.

  2. Unveiling the Backdoor Mechanism Hidden Behind Catastrophic Overfitting in Fast Adversarial Training

    cs.LG 2026-04 unverdicted novelty 5.0

    Catastrophic overfitting in fast adversarial training is reinterpreted as a weak-trigger variant of unlearnable tasks, allowing backdoor-inspired recalibration and outlier suppression to restore robustness.

  3. Backdoor Attacks and Defenses in Computer Vision Domain: A Survey

    cs.CR 2025-09 conditional novelty 3.0

    A literature review that classifies backdoor attacks and defenses in computer vision into a five-axis taxonomy and identifies supply-chain, hardware, and cross-task evaluation as open gaps.