Metraipolis › archive › dc_be3adc835c46

HASTE: Evolving Agent Harnesses Against Emerging Attacks Using Sparse Evidence — arXiv:2610.02920

Source
arXiv
Original
https://arxiv.org/abs/2610.02920
License (as archived)
CC0-1.0
Archived
2026-10-05T22:49:21.853Z
SHA-256 of the archived text
dc5d23f7ad449b691e435e4e186e056ea677f35ea50b55a075a80e8cbdb30756
Stable URL of this page
https://metraipolis.com/documento/dc_be3adc835c46

How to cite

HASTE: Evolving Agent Harnesses Against Emerging Attacks Using Sparse Evidence — arXiv:2610.02920 — arXiv, https://arxiv.org/abs/2610.02920 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_be3adc835c46 (Metraipolis archive dc_be3adc835c46, sha256 dc5d23f7ad449b691e435e4e186e056ea677f35ea50b55a075a80e8cbdb30756)
BibTeX
@misc{dcbe3adc835c46,
  title = {HASTE: Evolving Agent Harnesses Against Emerging Attacks Using Sparse Evidence — arXiv:2610.02920},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.02920},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_be3adc835c46; sha256 dc5d23f7ad449b691e435e4e186e056ea677f35ea50b55a075a80e8cbdb30756},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.02920. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Agent harnesses play a critical role in defenses by enforcing safety constraints to prevent unsafe actions. However, rapidly emerging attacks outpace manual harness adaptation, motivating automated harness evolution. Yet the signals available for harness evolution are often sparse, such as brief descriptions or a few attack examples in threat reports and preprints. To address this limitation, we introduce HASTE, a multi-agent framework that evolves agent harnesses from sparse threat evidence through an adversarial interplay between safety-specification generation and attack-case generation. Safety specifications guide harness updates toward addressing identified safety vulnerabilities, while attack cases probe for remaining safety vulnerabilities after each update. By feeding evaluation outcomes back into both processes, HASTE enables harness evolution against emerging attacks beyond the initially observed evidence. Experimental results across multiple backbone models, attack types, and evidence forms show that HASTE consistently reduces attack success rates while preserving benign-task utility. The code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.02920 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-05T22:45:55.651Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).