1 code implementation • 25 Mar 2024 • Reza Esfandiarpoor, Cristina Menghini, Stephen H. Bach
EX2 uses reinforcement learning to align a large language model with VLM preferences and generates descriptions that incorporate the important features for the VLM.
2 code implementations • 21 Feb 2024 • Zheng-Xin Yong, Cristina Menghini, Stephen H. Bach
We show that conditioning on bilingual lexicons is the key component of LexC-Gen. LexC-Gen is also practical -- it only needs a single GPU to generate data at scale.
no code implementations • 3 Oct 2023 • Zheng-Xin Yong, Cristina Menghini, Stephen H. Bach
AI safety training and red-teaming of large language models (LLMs) are measures to mitigate the generation of unsafe content.
2 code implementations • NeurIPS 2023 • Cristina Menghini, Andrew Delworth, Stephen H. Bach
We find that (1) unexplored prompt tuning strategies that iteratively refine pseudolabels consistently improve CLIP accuracy, by 19. 5 points in semi-supervised learning, by 28. 4 points in transductive zero-shot learning, and by 15. 2 points in unsupervised learning, and (2) unlike conventional semi-supervised pseudolabeling, which exacerbates model biases toward classes with higher-quality pseudolabels, prompt tuning leads to a more equitable distribution of per-class accuracy.
1 code implementation • 25 May 2022 • Alessio Mazzetto, Cristina Menghini, Andrew Yuan, Eli Upfal, Stephen H. Bach
We develop the first non-trivial lower bound on the worst-case error of the best map from attributes to classes for this setting, even with perfect attribute detectors.
2 code implementations • 8 Nov 2021 • Wasu Piriyakulkij, Cristina Menghini, Ross Briden, Nihal V. Nayak, Jeffrey Zhu, Elaheh Raisi, Stephen H. Bach
Machine learning practitioners often have access to a spectrum of data: labeled data for the target task (which is often limited), unlabeled data, and auxiliary data, the many available labeled datasets for other tasks.
no code implementations • 31 May 2019 • Aris Anagnostopoulos, Luca Becchetti, Adriano Fazzone, Cristina Menghini, Chris Schwiegelshohn
Reducing hidden bias in the data and ensuring fairness in algorithmic data analysis has recently received significant attention.