# Automatic Attack Discovery for Few-Shot Class-Incremental Learning via Large Language Models

arXiv: https://arxiv.org/abs/2512.03882 · HTML: https://mlsanity.com/paper/2512.03882/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 24 references that are in the map.
- First posted on arXiv 3 Dec 2025.

## Builds on (most cited first, 16 of 16)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Towards Evaluating the Robustness of Neural Networks](https://mlsanity.com/paper/1608.04644/) (2016)
- [Matching Networks for One Shot Learning](https://mlsanity.com/paper/1606.04080/) (2016)
- [DeepFool: a simple and accurate method to fool deep neural networks](https://mlsanity.com/paper/1511.04599/) (2015)
- [Few-Shot Class-Incremental Learning by Sampling Multi-Phase Tasks](https://mlsanity.com/paper/2203.17030/) (2022)
- [Continual Learning: Applications and the Road Forward](https://mlsanity.com/paper/2311.11908/) (2023)
- [Few-Shot Adversarial Prompt Learning on Vision-Language Models](https://mlsanity.com/paper/2403.14774/) (2024)
- [Learning Domain Invariant Prompt for Vision-Language Models](https://mlsanity.com/paper/2212.04196/) (2022)
- [OrCo: Towards Better Generalization via Orthogonality and Contrast for Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2403.18550/) (2024)
- [CLOSER: Towards Better Representation Learning for Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2410.05627/) (2024)
- [An Efficient Memory Module for Graph Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2411.06659/) (2024)
- [Adversarial Robustness of Prompt-based Few-Shot Learning for Natural Language Understanding](https://mlsanity.com/paper/2306.11066/) (2023)

## Abstract (arXiv)

Few-shot class incremental learning (FSCIL) is a more realistic and challenging paradigm in continual learning to incrementally learn unseen classes and overcome catastrophic forgetting on base classes with only a few training examples. Previous efforts have primarily centered around studying more effective FSCIL approaches. By contrast, less attention was devoted to thinking the security issues in contributing to FSCIL. This paper aims to provide a holistic study of the impact of attacks on FSCIL. We first derive insights by systematically exploring how human expert-designed attack methods (i.e., PGD, FGSM) affect FSCIL. We find that those methods either fail to attack base classes, or suffer from huge labor costs due to relying on huge expert knowledge. This highlights the need to craft a specialized attack method for FSCIL. Grounded in these insights, in this paper, we propose a simple yet effective ACraft method to automatically steer and discover optimal attack methods targeted at FSCIL by leveraging Large Language Models (LLMs) without human experts. Moreover, to improve the reasoning between LLMs and FSCIL, we introduce a novel Proximal Policy Optimization (PPO) based reinforcement learning to optimize learning, making LLMs generate better attack methods in the next generation by establishing positive feedback. Experiments on mainstream benchmarks show that our ACraft significantly degrades the performance of state-of-the-art FSCIL methods and dramatically beyond human expert-designed attack methods while maintaining the lowest costs of attack.