Rapsai: Accelerating Machine Learning Prototyping of Multimedia Applications through Visual Programming

Du, Ruofei; Li, Na; Jin, Jing; Carney, Michelle Mohr; Miles, Scott; Kleiner, Maria; Yuan, Xiuxiu; Zhang, Yinda; Kulkarni, Anuva; Liu, Xingyu; Sabie, Ahmed; Orts-Escolano, Sergio; Kar, Abhishek; Ping, Yu; Iyengar, Ram; Kowdle, Adarsh; Olwal, Alex

doi:10.1145/3544548.3581338

Cited by 16 publications

(1 citation statement)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Layer-wise relevance propagation decomposed the model's output into contributions from each layer, offering a detailed view of how information flowed through the network and how each component contributed to the final decision [25,26]. Human-model interaction studies indicated that users found models with interpretable explanations more trustworthy and easier to use, enhancing their overall effectiveness in practical applications [27]. Usability studies emphasized the importance of providing clear and concise explanations, which significantly improved user satisfaction and decision-making efficiency [28].…”

Section: Introductionmentioning

confidence: 99%

An Evaluation of the Safety of ChatGPT with Malicious Prompt Injection

Han,

Guo

2024

Preprint

View full text Add to dashboard Cite

Artificial intelligence systems, particularly those involving sophisticated neural network architectures like ChatGPT, have demonstrated remarkable capabilities in generating human-like text. However, the susceptibility of these systems to malicious prompt injections poses significant risks, necessitating comprehensive evaluations of their safety and robustness. The study presents a novel automated framework for systematically injecting and analyzing malicious prompts to assess the vulnerabilities of ChatGPT. Results indicate substantial rates of harmful responses across various scenarios, highlighting critical areas for improvement in model defenses. The findings underscore the importance of advanced adversarial training, real-time monitoring, and interdisciplinary collaboration to enhance the ethical deployment of AI systems. Recommendations for future research emphasize the need for robust safety mechanisms and transparent model operations to mitigate the risks associated with adversarial inputs.

show abstract

Section: Introductionmentioning

confidence: 99%