Paper Detail

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

Tongyan Hu, Hao Li, Xiaogeng Liu, Ruida Wang, Zhengyu Liu, Shuyao Xu, Ning Zhang, Ziyang Li, Yinzhi Cao, Bryan Hooi, Chaowei Xiao

arxiv Score 7.8

Published 2026-10-04 · First seen 2026-10-06

General AI

Abstract

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 https://github.com/SaFo-Lab/Red-TTT

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{hu2026red,
  title = {Red-TTT: Test-Time Training for Automated Jailbreaking Large Language Models},
  author = {Tongyan Hu and Hao Li and Xiaogeng Liu and Ruida Wang and Zhengyu Liu and Shuyao Xu and Ning Zhang and Ziyang Li and Yinzhi Cao and Bryan Hooi and Chaowei Xiao},
  year = {2026},
  abstract = {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 win},
  url = {https://arxiv.org/abs/2610.05282},
  keywords = {cs.CL, cs.CR},
  eprint = {2610.05282},
  archiveprefix = {arXiv},
}

Metadata

{}