← Back to all news
arXiv AI By Indranil Halder, Annesya Banerjee, Cengiz Pehlevan

Jailbreak Scaling Laws for Large Language Models: Polynomial-Exponential Crossover

Read the original on arXiv AI →

The paper investigates how adversarial prompt‑injection attacks can shift the scaling of attack success from a slow polynomial growth to an exponential growth as the number of inference‑time samples increases. By proposing a minimal statistical framework and a spin‑glass generative model, the authors explain that short injected prompts act as a weak magnetic field leading to power‑law scaling, while long prompts act as a strong field producing exponential scaling. The phenomenon is observed consistently across large language models ranging from 3B to 70B parameters and across various attack methods and benchmark datasets.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Machine Learning
Oct 2

Inference-Layer Security: Defending Against Adversarial Inference and Infrastructure Abuse

The paper presents a study on securing large language model (LLM) inference services against adversarial use, such as jailbreaking, denial‑of‑service, and distillation attacks. It introduces a structural causal model to generate a realistic, labeled dataset of user sessions, including coordinated multi‑account campaigns and varying label observability. Using this dataset, the authors train a gradient‑boosted detector that achieves near‑perfect binary classification on oracle labels but shows lower performance on operational labels, and they demonstrate a simple thresholded engine that improves attack‑type attribution accuracy.

By Keifer Lee
llmsefficiencysafety
More like this →
arXiv AI
Jul 9

NonTextual Target Attack

arXiv:2510. 02999v5 Announce Type: replace-cross Abstract: Existing gradient-based jailbreak attacks on Large Language Models (LLMs) typically optimize adversarial suffixes to align the LLM output with predefined target responses.

By Xinzhe Huang, Wenjing Hu, Tianhang Zheng, Kedong Xiu, Hongsheng Hu, Xiaojun Jia, Di Wang, Zhan Qin, Kui Ren
llmsbenchmarkssafety
More like this →
arXiv AI
Aug 26

ADVERSA: Measuring Multi-Turn Guardrail Degradation and Judge Reliability in Large Language Models

The paper introduces ADVERSA, an automated red‑teaming framework that evaluates large language model safety over multiple turns by tracking continuous compliance trajectories instead of binary jailbreak outcomes. Using a fine‑tuned 70B attacker model and a structured 5‑point rubric, the authors conduct controlled experiments on three frontier victim models, measuring guardrail degradation and judge reliability through a triple‑judge consensus. Results show a 26.7% jailbreak rate with most breaches occurring early, and the study documents inter‑judge agreement, attacker drift, and attacker refusals as key factors affecting safety assessment.

By Harry Owiredu-Ashley
llmsfine-tuningsafety
More like this →