Metraipolis › archive › dc_14befe347c26

Red-TTT: Test-Time Training for Automated Jailbreaking Large Language Models — arXiv:2610.05282

Source
arXiv
Original
https://arxiv.org/abs/2610.05282
License (as archived)
CC0-1.0
Archived
2026-10-06T05:28:18.654Z
SHA-256 of the archived text
9d700f0d159177f6045a1f24ade04f673e619d79fe2dc229fef0aa3195191df5
Stable URL of this page
https://metraipolis.com/documento/dc_14befe347c26

How to cite

Red-TTT: Test-Time Training for Automated Jailbreaking Large Language Models — arXiv:2610.05282 — arXiv, https://arxiv.org/abs/2610.05282 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_14befe347c26 (Metraipolis archive dc_14befe347c26, sha256 9d700f0d159177f6045a1f24ade04f673e619d79fe2dc229fef0aa3195191df5)
BibTeX
@misc{dc14befe347c26,
  title = {Red-TTT: Test-Time Training for Automated Jailbreaking Large Language Models — arXiv:2610.05282},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.05282},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_14befe347c26; sha256 9d700f0d159177f6045a1f24ade04f673e619d79fe2dc229fef0aa3195191df5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05282. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models remain vulnerable to jailbreaks, and automated red teaming is the standard way to find jailbreaks in large language models at scale. Current methods either draw more samples at test time through search, rewriting, and tree expansion, or train a stronger attacker offline with reinforcement learning. Both share a limitation: once an attack on a specific target behavior begins, the attacker's weights are frozen. Any signal it gathers about the behavior stays in its context window and is discarded afterward. The attacker never adapts its proposal distribution mid-attack, so success depends almost entirely on the sampling budget, and under a budget affordable at scale, many behaviors remain unbroken. We propose Red-TTT, which updates the attacker's parameters during the attack on each behavior. At each round, the attacker samples a group of candidates, scores them against the victim's replies, and takes a policy-gradient step before drawing the next group, so what it discovers about the current victim is consolidated into weights rather than accumulated as context. We also adapt the training objective to red teaming, where success is judged by the single best sample rather than the average. Red-TTT requires only sampling access to the victim and integrates into existing attack pipelines with no other changes. Against the Best-of-N baseline, Red-TTT raises attack success rate from 55.9\% to 72.4\% on average at a budget of 120 samples, improving over the baseline in every configuration and cracking many behaviors previous method cannot. The code is available at this https URL

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2610.05282 — 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 2026-10-06T05:24:20.322Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).