27 papers
Over-Refusal and Representation Subspaces: A Mechanistic Analysis of Task-Conditioned Refusal in Aligned LLMs
Utsav Maskey, Mark Dras, Usman Naseem
Aligned language models that are trained to refuse harmful requests also exhibit over-refusal: they decline safe instructions that seemingly resemble harmful instructions. A natura…
We Think, Therefore We Align LLMs to Helpful, Harmless and Honest Before They Go Wrong
Gautam Siddharth Kashyap, Mark Dras, Usman Naseem
Alignment of Large Language Models (LLMs) is the ability to satisfy desired objectives during generation, which is critical for trustworthy deployment. In practice, alignment is of…
Fairness Evaluation and Inference Level Mitigation in LLMs
Afrozah Nadeem, Mark Dras, Usman Naseem
Large language models often display undesirable behaviors embedded in their internal representations, undermining fairness, inconsistency drift, amplification of harmful content, a…
AlignCultura: Towards Culturally Aligned Large Language Models?
Gautam Siddharth Kashyap, Mark Dras, Usman Naseem
Cultural alignment in Large Language Models (LLMs) is essential for producing contextually aware, respectful, and trustworthy outputs. Without it, models risk generating stereotype…
SafeConstellations: Mitigating Over-Refusals in LLMs Through Task-Aware Representation Steering
Utsav Maskey, Sumit Yadav, Mark Dras +1
LLMs increasingly exhibit over-refusal behavior, where safety mechanisms cause models to reject benign instructions that seemingly resemble harmful content. This phenomenon diminis…
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…