paper-with-me

Papers

Harnessing non-adversarial robustness in large language models

2026-05-28 · Qinghua Zhou, Ellina Aleshina, Andrey Lovyagin, Oleg Somov, Mikhail Seleznyov, Alexander Panchenko, Ivan Oseledets, Elena Tutubalina, Ivan Y. Tyukin arxiv

The work presents an approach for addressing the challenge of robustness in Large Language Models (LLMs) to alterations and potential errors caused by semantically similar but textually different prompts. Recent works have shown that these kinds of prompt variations can significantly impact the performance of LLMs on tasks. The central question is: can LLMs' robustness to semantically-neutral prompt alterations be acquired without expensive retraining of the entire model? We address this question both theoretically and through experiments. Our theoretical analysis reveals a crucial factor impacting model robustness - a systematic expected shift or perturbation-induced bias in neural network module outputs. Motivated by this analysis, we show that robustness can be achieved via a simple fine-tuning process: debiasing for robustness. We identify conditions when debiasing helps and when it does not, and demonstrate, through both theory and extensive experiments, that debiasing for robustness may indeed be a quick and efficient tool to enhance robustness and provide certification against random prompt perturbations.

📄 PDF Abstract BibTeX arXiv:2605.29816

Code (0)

등록된 구현이 없습니다.

Tasks

Adversarial Robustness

Similar Papers 제목 키워드 기반

Adapters Mixup: Mixing Parameter-Efficient Adapters to Enhance the Adversarial Robustness of Fine-tuned Pre-trained Text Classifiers

2024-01-18 · Tuc Nguyen, Thai Le

Existing works show that augmenting the training data of pre-trained language models (PLMs) for classification tasks fine-tuned via parameter-efficient fine-tuning methods (PEFT) using both clean and adversarial examples…

Adversarial Robustnessparameter-efficient fine-tuningtext-classificationText Classification

Harnessing the Power of Adversarial Prompting and Large Language Models for Robust Hypothesis Generation in Astronomy

2023-06-20 · Ioana Ciucă, Yuan-Sen Ting, Sandor Kruk, Kartheik Iyer

This study investigates the application of Large Language Models (LLMs), specifically GPT-4, within Astronomy. We employ in-context prompting, supplying the model with up to 1000 papers from the NASA Astrophysics Data Sy…

Astronomy

Stable and Efficient Adversarial Training through Local Linearization

2022-10-11 · Zhuorong Li, Daiwei Yu

There has been a recent surge in single-step adversarial training as it shows robustness and efficiency. However, a phenomenon referred to as ``catastrophic overfitting" has been observed, which is prevalent in single-st…

Harnessing Perceptual Adversarial Patches for Crowd Counting

2021-09-16 · Shunchang Liu, Jiakai Wang, Aishan Liu, Yingwei Li 외

Crowd counting, which has been widely adopted for estimating the number of people in safety-critical scenes, is shown to be vulnerable to adversarial examples in the physical world (e.g., adversarial patches). Though har…

Crowd Counting

Harnessing the Vulnerability of Latent Layers in Adversarially Trained Models

2019-05-13 · Mayank Singh, Abhishek Sinha, Nupur Kumari, Harshitha Machiraju 외

Neural networks are vulnerable to adversarial attacks -- small visually imperceptible crafted noise which when added to the input drastically changes the output. The most effective method of defending against these adver…

Adversarial Attack