13 papers
Pragmatic Attack Surface: Vulnerabilities of Implicit Context in Large Language Models
Bocheng Chen, Han Zi, Roucheng Ou +5
In the era of large language models (LLMs), attackers often manipulate natural language to elicit unsafe or harmful outputs, creating a new natural language attack surface unique t…
Deactivating Refusal Triggers: Understanding and Mitigating Overrefusal in Safety Alignment
Zhiyu Xue, Zimo Qi, Guangliang Liu +2
Safety alignment aims to ensure that large language models (LLMs) refuse harmful requests by post-training on harmful queries paired with refusal answers. Although safety alignment…
Self-correction is Not An Innate Capability in Language Models
Guangliang Liu, Zimo Qi, Xitong Zhang +2
Although there has been growing interest in the self-correction capability of Large Language Models (LLMs), there are varying conclusions about its effectiveness. Prior research ha…
Diagnosing the Performance Trade-off in Moral Alignment: A Case Study on Gender Stereotypes
Guangliang Liu, Bocheng Chen, Han Zi +2
Moral alignment has emerged as a widely adopted approach for regulating the behavior of pretrained language models (PLMs), typically through fine-tuning on curated datasets. Gender…
Discourse Heuristics For Paradoxically Moral Self-Correction
Guangliang Liu, Zimo Qi, Xitong Zhang +1
Moral self-correction has emerged as a promising approach for aligning the output of Large Language Models (LLMs) with human moral values. However, moral self-correction techniques…
On the Convergence of Moral Self-Correction in Large Language Models
Guangliang Liu, Haitao Mao, Bochuan Cao +4
Large Language Models (LLMs) are able to improve their responses when instructed to do so, a capability known as self-correction. When instructions provide only a general and abstr…