Chain-of-Thought Driven Adversarial Scenario Extrapolation for Robust Language Models
Md Rafi Ur Rashid, Vishnu Asutosh Dasu, Ye Wang, Gang Tan, Shagufta Mehnaz
Abstract
Large Language Models (LLMs) exhibit impressive capabilities, but remain susceptible to a growing spectrum of safety risks, including jailbreaks, toxic content, hallucinations, and bias. Existing defenses often address only a single threat type or resort to rigid outright rejection, sacrificing user experience and failing to generalize across diverse and novel attacks. This paper introduces Adversarial Scenario Extrapolation (ASE), a novel inference-time computation framework that leverages Chain-of-Thought (CoT) reasoning to simultaneously enhance LLM robustness and seamlessness. ASE guides the LLM through a self-generative process of contemplating potential adversarial scenarios and formulating defensive strategies before generating a response to the user query. Comprehensive evaluation on four adversarial benchmarks with four latest LLMs shows that ASE achieves near-zero jailbreak attack success rates and minimal toxicity, while slashing outright rejections to
BibTeX
@inproceedings{aaai2026_chainofthoughtdr,
title = {Chain-of-Thought Driven Adversarial Scenario Extrapolation for Robust Language Models},
author = {Md Rafi Ur Rashid and Vishnu Asutosh Dasu and Ye Wang and Gang Tan and Shagufta Mehnaz},
booktitle = {AAAI 2026},
year = {2026}
}