Urgent.News

What's breaking now, across thousands of outlets.

AI

Evaluating in-context learning with prompting regimes for transformer-based synthetic health tabular data generation

Tabular synthetic data generation (SDG) can facilitate research and development in the health domain, as access to health data is restricted. To accommodate for its structural complexity, such as multimodal distributions and fragmentation, transformer models provides an alternative to conventional SDG methods since they are better equipped to infer global spatial coherence due to textual context…

Tabular synthetic data generation (SDG) has the potential to aid research and development in the health sector, as access to health data is often restricted. Due to the intricate nature of such data, transformer models have emerged as a more suitable alternative to traditional SDG techniques. These models excel at capturing global spatial coherence within tabular records through the utilization of textual context.

However, the optimal strategies for adapting transformer models to SDG tasks remain largely unexplored. This study aims to shed light on this matter by examining the fidelity and privacy aspects of three prompting regimes, namely zero shot, one shot, and few shot training, applied to tabular health SDG using compact transformer models.

The research team assessed the performance of llama 3.2 (1B and 3B parameters) and Phi (1.3B and 2.7B parameters) on a real clinical dataset. The findings revealed that zero shot learning exhibited a lower propensity for hallucinations and generally outperformed in terms of fidelity. Nevertheless, this advantage came at the expense of privacy, as compared to the alternative prompting regimes.

Interestingly, the study observed no discernible differences in performance across the various model sizes and families tested. The researchers suggest that future investigations should broaden the scope of this analysis to encompass additional models and datasets, as well as investigate potential methods to mitigate the presence of hallucinations.

Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at biorxiv.org →

More in AI

60 Minutes Transcript: Artificial intelligence and the future of work

Some Americans are working to improve artificial intelligence, teaching it the skills and knowledge accrued over the course of a career.

  • AI seen as primary reason for U.S. company layoffs in past year
  • Mercor grows from 20 to 400 employees, hires 100k freelancers
  • AI augments human expertise, making workers more productive

Experts are training AI how to do their jobs

It's become a booming industry: teaching artificial intelligence how to get better at doing all kinds of jobs. As AI advances, many fear what this tech means for the future of work.

  • AI trained on industry experts to augment human jobs
  • 75% of Americans fear AI job displacement, despite optimism
  • AI personas assist firms like Vorys in document preparation

Hola, hang tight go check it out (Hawk) AI powered panel with multiple models

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend What I Built Hawk is a debate room for AI models.

  • Hawk is an AI-powered panel facilitating multi-model debates for coding tasks
  • Users summon three AI models in separate terminal-style windows to address questions
  • App is free, runs on Render's free tier, supports multiple models from various providers

More from Monday 5 October →