67 citations · 182 across the 41 of their papers we have counts for
42 papers
Segment-Level Coherence for Robust Harmful Intent Probing in LLMs
Xuanli He, Bilgehan Sel, Faizan Ali +3
Large Language Models (LLMs) are increasingly exposed to adaptive jailbreaking, particularly in high-stakes Chemical, Biological, Radiological, and Nuclear (CBRN) domains. Although…
Trojan-Speak: Bypassing Constitutional Classifiers with No Jailbreak Tax via Adversarial Finetuning
Bilgehan Sel, Xuanli He, Alwin Peng +2
Fine-tuning APIs offered by major AI providers create new attack surfaces where adversaries can bypass safety measures through targeted fine-tuning. We introduce Trojan-Speak, an a…
From KMMLU-Redux to KMMLU-Pro: A Professional Korean Benchmark Suite for LLM Evaluation
Seokhee Hong, Sunkyoung Kim, Guijin Son +3
The development of Large Language Models (LLMs) requires robust benchmarks that encompass not only academic domains but also industrial fields to effectively evaluate their applica…
Defending against Backdoor Attacks via Module Switching
Weijun Li, Ansh Arora, Xuanli He +2
Backdoor attacks pose a serious threat to deep neural networks (DNNs), allowing adversaries to implant triggers for hidden behaviors in inference. Defending against such vulnerabil…
Cut the Deadwood Out: Backdoor Purification via Guided Module Substitution
Yao Tong, Weijun Li, Xuanli He +2
Model NLP models are commonly trained (or fine-tuned) on datasets from untrusted platforms like HuggingFace, posing significant risks of data poisoning attacks. A practical yet und…
Analysing the Residual Stream of Language Models Under Knowledge Conflicts
Yu Zhao, Xiaotang Du, Giwon Hong +6
Large language models (LLMs) can store a significant amount of factual knowledge in their parameters. However, their parametric knowledge may conflict with the information provided…
Are We Done with MMLU?
Aryo Pradipta Gema, Joshua Ong Jun Leang, Giwon Hong +13
Maybe not. We identify and analyse errors in the popular Massive Multitask Language Understanding (MMLU) benchmark. Even though MMLU is widely adopted, our analysis demonstrates nu…
IrokoBench: A New Benchmark for African Languages in the Age of Large Language Models
David Ifeoluwa Adelani, Jessica Ojo, Israel Abebe Azime +24
Despite the widespread adoption of Large language models (LLMs), their remarkable capabilities remain limited to a few high-resource languages. Additionally, many low-resource lang…
SEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks
Xuanli He, Qiongkai Xu, Jun Wang +2
Modern NLP models are often trained on public datasets drawn from diverse sources, rendering them vulnerable to data poisoning attacks. These attacks can manipulate the model's beh…
Attacks on Third-Party APIs of Large Language Models
Wanru Zhao, Vidit Khazanchi, Haodi Xing +3
Large language model (LLM) services have recently begun offering a plugin ecosystem to interact with third-party API services. This innovation enhances the capabilities of LLMs, bu…
The Hallucinations Leaderboard -- An Open Effort to Measure Hallucinations in Large Language Models
Giwon Hong, Aryo Pradipta Gema, Rohit Saxena +8
Large Language Models (LLMs) have transformed the Natural Language Processing (NLP) landscape with their remarkable ability to understand and generate human-like text. However, the…
Backdoor Attack on Multilingual Machine Translation
Jun Wang, Qiongkai Xu, Xuanli He +2
While multilingual machine translation (MNMT) systems hold substantial promise, they also have security vulnerabilities. Our research highlights that MNMT systems can be susceptibl…
TuBA: Cross-Lingual Transferability of Backdoor Attacks in LLMs with Instruction Tuning
Xuanli He, Jun Wang, Qiongkai Xu +4
The implications of backdoor attacks on English-centric large language models (LLMs) have been widely examined - such attacks can be achieved by embedding malicious behaviors durin…
StarCoder 2 and The Stack v2: The Next Generation
Anton Lozhkov, Raymond Li, Loubna Ben Allal +63
The BigCode project, an open-scientific collaboration focused on the responsible development of Large Language Models for Code (Code LLMs), introduces StarCoder2. In partnership wi…
Generative Models are Self-Watermarked: Declaring Model Authentication through Re-Generation
Aditya Desu, Xuanli He, Qiongkai Xu +1
As machine- and AI-generated content proliferates, protecting the intellectual property of generative models has become imperative, yet verifying data ownership poses formidable ch…
Here's a Free Lunch: Sanitizing Backdoored Models with Model Merge
Ansh Arora, Xuanli He, Maximilian Mozes +3
The democratization of pre-trained language models through open-source initiatives has rapidly advanced innovation and expanded access to cutting-edge technologies. However, this o…
AfriMTE and AfriCOMET: Enhancing COMET to Embrace Under-resourced African Languages
Jiayi Wang, David Ifeoluwa Adelani, Sweta Agrawal +55
Despite the recent progress on scaling multilingual machine translation (MT) to several under-resourced African languages, accurately measuring this progress remains challenging, s…
Using Natural Language Explanations to Improve Robustness of In-context Learning
Xuanli He, Yuxiang Wu, Oana-Maria Camburu +2
Recent studies demonstrated that large language models (LLMs) can excel in many tasks via in-context learning (ICL). However, recent works show that ICL-prompted models tend to pro…
Use of LLMs for Illicit Purposes: Threats, Prevention Measures, and Vulnerabilities
Maximilian Mozes, Xuanli He, Bennett Kleinberg +1
Spurred by the recent rapid increase in the development and distribution of large language models (LLMs) across industry and academia, much recent work has drawn attention to safet…
Can Knowledge Graphs Simplify Text?
Anthony Colas, Haodi Ma, Xuanli He +2
Knowledge Graph (KG)-to-Text Generation has seen recent improvements in generating fluent and informative sentences which describe a given KG. As KGs are widespread across multiple…
IMBERT: Making BERT Immune to Insertion-based Backdoor Attacks
Xuanli He, Jun Wang, Benjamin Rubinstein +1
Backdoor attacks are an insidious security threat against machine learning models. Adversaries can manipulate the predictions of compromised models by inserting triggers into the t…
G3Detector: General GPT-Generated Text Detector
Haolan Zhan, Xuanli He, Qiongkai Xu +2
The burgeoning progress in the field of Large Language Models (LLMs) heralds significant benefits due to their unparalleled capacities. However, it is critical to acknowledge the p…
Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation
Xuanli He, Qiongkai Xu, Jun Wang +2
Modern NLP models are often trained over large untrusted datasets, raising the potential for a malicious adversary to compromise model behaviour. For instance, backdoors can be imp…
Koala: An Index for Quantifying Overlaps with Pre-training Corpora
Thuy-Trang Vu, Xuanli He, Gholamreza Haffari +1
In very recent years more attention has been placed on probing the role of pre-training data in Large Language Models (LLMs) downstream behaviour. Despite the importance, there is…