Can a large language model be a gaslighter?

Large language models (LLMs) have gained human trust due to their capabilities and helpfulness. However, this in turn may allow LLMs to affect users' mindsets by manipulating language. It is termed as gaslighting, a psychological effect. In this work, we aim to investigate the vulnerability of LLMs under prompt-based and fine-tuning-based gaslighting attacks. Therefore, we propose a two-stage framework DeepCoG designed to: 1) elicit gaslighting plans from LLMs with the proposed DeepGaslighting prompting template, and 2) acquire gaslighting conversations from LLMs through our Chain-of-Gaslighting method. The gaslighting conversation dataset along with a corresponding safe dataset is applied to fine-tuning-based attacks on open-source LLMs and anti-gaslighting safety alignment on these LLMs. Experiments demonstrate that both prompt-based and fine-tuning-based attacks transform three open-source LLMs into gaslighters. In contrast, we advanced three safety alignment strategies to strengthen (by 12.05%) the safety guardrail of LLMs. Our safety alignment strategies have minimal impacts on the utility of LLMs. Empirical studies indicate that an LLM may be a potential gaslighter, even if it passed the harmfulness test on general dangerous queries.

Review of photoacoustic imaging plus X

Photoacoustic imaging (PAI) is a novel modality in biomedical imaging technology that combines the rich optical contrast with the deep penetration of ultrasound. To date, PAI technology has found applications in various biomedical fields. In this review, we present an overview of the emerging research frontiers on PAI plus other advanced technologies, named as PAI plus X, which includes but not limited to PAI plus treatment, PAI plus new circuits design, PAI plus accurate positioning system, PAI plus fast scanning systems, PAI plus novel ultrasound sensors, PAI plus advanced laser sources, PAI plus deep learning, and PAI plus other imaging modalities. We will discuss each technology's current state, technical advantages, and prospects for application, reported mostly in recent three years. Lastly, we discuss and summarize the challenges and potential future work in PAI plus X area.