Dataset Condensation Atlas

Trustworthy DD

Attention Hijacking

Attention Hijacking: Backdooring Text Dataset Distillation via Semantic Anchors

Hang Ren

ICML 2026 · first public 2026-01-01

paper ↗catalogued✓ metadata verified

In one paragraph

Proposes a backdoor attack on text dataset distillation built on a "Semantic Anchoring Hypothesis": the attack reshapes gradients into input embeddings so the synthetic data evolves to turn a trigger word into an adversarial feature, integrated into the bi-level distillation loop so the attack satisfies both the clean-task and backdoor objectives at once.

Where it sits

BibTeX (generated; prefer the venue's official entry)
@article{ren2026attention,
  title   = {Attention Hijacking: Backdooring Text Dataset Distillation via Semantic Anchors},
  author  = {Hang Ren},
  journal = {ICML 2026},
  year    = {2026}
}