Endovascular treatment of large hemoptysis for pulmonary pseudoaneurysm: Report of 23 cases

Li, Fenqiang; Su, Dong-Jun; Zhang, Wanjia; Chen, Zhong-Ke; Li, Gengxiang; Li, Shuangxi; Peng, Yuxing; Dang, Lei; Wang, Wenhui

doi:10.21203/rs.3.rs-2400821/v1

Cited by 2 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, it is essential to carefully evaluate the model prediction performance and explanation quality. Other than exploring more advanced ML algorithms, LLMs [8,47,84,88] may offer a new method to generate appropriate and convincing explanation content. Besides, when an intervention is personalized, users, especially younger ones, could be biased towards being more receptive to adopting it [24].…”

Section: Ethical Concerns and Risk Of Ai-based Intervention Systemmentioning

confidence: 99%

Time2Stop: Adaptive and Explainable Human-AI Loop for Smartphone Overuse Intervention

Orzikulova,

Xiao,

et al. 2024

Proceedings of the CHI Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

Despite a rich history of investigating smartphone overuse intervention techniques, AI-based just-in-time adaptive intervention (JITAI) methods for overuse reduction are lacking. We develop Time2Stop, an intelligent, adaptive, and explainable JITAI system that leverages machine learning to identify optimal intervention timings, introduces interventions with transparent AI explanations, and collects user feedback to establish a human-AI loop and adapt the intervention model over time. We conducted an 8-week field experiment (N=71) to evaluate the effectiveness of both the adaptation and explanation aspects of Time2Stop. Our results indicate that our adaptive models significantly outperform the baseline methods on intervention accuracy (>32.8% relatively) and receptivity (>8.0%). In addition, incorporating explanations further enhances the effectiveness by 53.8% and 11.4% on accuracy and receptivity, respectively. Moreover, Time2Stop significantly reduces overuse, decreasing app visit frequency by 7.0∼8.9%. Our subjective data also echoed these quantitative measures. Participants preferred the adaptive interventions and rated the system highly on intervention

show abstract

Section: Ethical Concerns and Risk Of Ai-based Intervention Systemmentioning

confidence: 99%

Time2Stop: Adaptive and Explainable Human-AI Loop for Smartphone Overuse Intervention

Orzikulova,

Xiao,

et al. 2024

Proceedings of the CHI Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

show abstract

“…Since the LLMs do not seem to share the same understanding of prompts with humans [49,69], a series of works in prompt engineering have been proposed [58,59,67,72]. Among them, one particular prompt strategy called Chain-of-thought (CoT) has been shown to elicit strong reasoning abilities in LLMs by asking the model to incorporate intermediate reasoning steps (rationales) while solving a problem [38,40,66,72]. Wang et al [67] sample from the reasoning path and vote for the majority result.…”

Section: In-context Learning and Chain-of-thoughtmentioning

confidence: 99%

“…Such reasoning ability is not a feature only found in LLMs, and several works have explored incorporating it in small models. Li et al [38] used CoT-like reasoning from LLMs to train smaller models on a joint task of generating the solution and explaining the solution generated. With multi-step reasoning, Fu et al [26] concentrate small models on a specific task and sacrifice its generalizability in exchange for the high performance.…”

Section: In-context Learning and Chain-of-thoughtmentioning

confidence: 99%

On the Feasibility of Specialized Ability Stealing for Large Language Code Models

Li¹,

Wang²,

Ma³

et al. 2023

Preprint

View full text Add to dashboard Cite

Recent progress in large language code models (LLCMs) has led to a dramatic surge in the use of software development. Nevertheless, it is widely known that training a well-performed LLCM requires a plethora of workforce for collecting the data and high quality annotation. Additionally, the training dataset may be proprietary (or partially open source to the public), and the training process is often conducted on a large-scale cluster of GPUs with high costs. Inspired by the recent success of imitation attacks in extracting computer vision and natural language models, this work launches the first imitation attack on LLCMs: by querying a target LLCM with carefully-designed queries and collecting the outputs, the adversary can train an imitation model that manifests close behavior with the target LLCM. We systematically investigate the effectiveness of launching imitation attacks under different query schemes and different LLCM tasks. We also design novel methods to polish the LLCM outputs, resulting in an effective imitation training process. We summarize our findings and provide lessons harvested in this study that can help better depict the attack surface of LLCMs. Our research contributes to the growing body of knowledge on imitation attacks and defenses in deep neural models, particularly in the domain of code related tasks.

show abstract

Endovascular treatment of large hemoptysis for pulmonary pseudoaneurysm: Report of 23 cases

Cited by 2 publications

References 25 publications

Time2Stop: Adaptive and Explainable Human-AI Loop for Smartphone Overuse Intervention

Time2Stop: Adaptive and Explainable Human-AI Loop for Smartphone Overuse Intervention

On the Feasibility of Specialized Ability Stealing for Large Language Code Models

Contact Info

Product

Resources

About