Project description
Background/Motivation:Backdoor attacks are attacks on neural networks where a so-called trigger alters the decision-making behaviour of the networks, thereby creating vulnerabilities. These triggers can be injected into the training dataset or directly into the model weights. These are then called poisoned. Due to parameter-efficient fine-tuning methods, backdoor attacks on large language models (LLMs) have become significantly more difficult to detect, as a poisoned parameter update is harder to recognise than a poisoned dataset. Therefore, several methods have been developed recently to detect poisoned model updates.Objective: Due to the variety of backdoor attacks, methods often detect far fewer attacks than they claim, as they frequently make assumptions that do not correspond to reality. The aim of this work is therefore to identify and exploit vulnerabilities in methods presented in the literature, so that the promised effects are not achieved as desired.Results: The results of this work are intended to demonstrate to the research community that a fundamental understanding of the mechanics of backdoor attacks is absolutely necessary. For this purpose, models, datasets, or finetuning adapters should be created and tested for previously selected detection methods, which can reliably bypass these detection methods. Be part of change Researching and implementing novel machine learning approaches that enhance the security of LLMs. Self-critical evaluation of the obtained results. Presenting the results. Preparing a project report in the form of a master's thesis.
Funding and compensation
Compensation not specified. See the original posting for amounts, duration and conditions.
Eligibility and application requirements
Review the qualifications, research interests and required documents in the description above. The employer or university's original posting is the source for complete eligibility requirements.
How to apply
Submit your application through Fraunhofer's official application process. Apply early as a closing date has not been specified.
Source and listing information
This opportunity is published by Fraunhofer. Read the original official posting. Conditions and availability may change; confirm them with the organization.
Report an outdated or incorrect listing