Прати
Dylan Sam
Dylan Sam
Верификована је имејл адреса на andrew.cmu.edu - Почетна страница
Наслов
Навело
Навело
Година
Adversarial Multiclass Learning under Weak Supervision with Performance Guarantees
A Mazzetto, C Cousins, D Sam, SH Bach, E Upfal
International Conference on Machine Learning, 2021
392021
Semi-supervised aggregation of dependent weak supervision sources with performance guarantees
A Mazzetto, D Sam, A Park, E Upfal, S Bach
International Conference on Artificial Intelligence and Statistics, 3196-3204, 2021
342021
Losses over labels: Weakly supervised learning via direct loss construction
D Sam, JZ Kolter
Proceedings of the AAAI Conference on Artificial Intelligence 37 (8), 9695-9703, 2023
122023
Understanding prompt engineering may not require rethinking generalization
V Akinwande, Y Jiang, D Sam, JZ Kolter
arXiv preprint arXiv:2310.03957, 2023
102023
Label Propagation with Weak Supervision
R Pukdee, D Sam, MF Balcan, P Ravikumar
International Conference on Learning Representations, 2023
102023
Bayesian neural networks with domain knowledge priors
D Sam, R Pukdee, DP Jeong, Y Byun, JZ Kolter
arXiv preprint arXiv:2402.13410, 2024
82024
Auditing fairness under unobserved confounding
Y Byun, D Sam, M Oberst, Z Lipton, B Wilder
International Conference on Artificial Intelligence and Statistics, 4339-4347, 2024
72024
Learning with explanation constraints
R Pukdee, D Sam, JZ Kolter, MF Balcan, P Ravikumar
arXiv preprint arXiv:2303.14496, 2023
72023
Automated data accountability for missions in Mars rover data
R Alimo, D Sam, D Lakhmiri, B Kahovec, D Divsalar
2021 IEEE Aerospace Conference (50100), 1-8, 2021
7*2021
Computing low-entropy couplings for large-support distributions
S Sokota, D Sam, CS de Witt, S Compton, J Foerster, JZ Kolter
arXiv preprint arXiv:2405.19540, 2024
12024
Learning from Dependent Weak Supervision Sources
D Sam
Brown University, 2021
1*2021
Analyzing Similarity Metrics for Data Selection for Language Model Pretraining
D Sam, A Chakrabarti, A Rostamizadeh, S Ramalingam, G Citovsky, ...
arXiv preprint arXiv:2502.02494, 2025
2025
Predicting the Performance of Black-box LLMs through Self-Queries
D Sam, M Finzi, JZ Kolter
arXiv preprint arXiv:2501.01558, 2025
2025
System and method for prompt searching
DT Willmott, VA Akinwande, Y Jiang, DJ Sam, J Kolter
US Patent App. 18/217,248, 2025
2025
Finetuning CLIP to Reason about Pairwise Differences
D Sam, D Willmott, JD Semedo, JZ Kolter
arXiv preprint arXiv:2409.09721, 2024
2024
Improving self-supervised representation learning via sequential adversarial masking
D Sam, M Bai, T McKinney, LE Li
Self-Supervised Learning: Theory and Practice @ NeurIPS, 2022, 2022
2022
The Promises and Pitfalls of Language Models for Structured Numerical Data
N Gruver, MA Finzi, D Sam, JZ Kolter, B Athiwaratkun, AG Wilson
Eliciting Black-Box Representations from LLMs through Self-Queries
D Sam, MA Finzi
ICML 2024 Next Generation of AI Safety Workshop, 0
Систем тренутно не може да изврши ову радњу. Пробајте поново касније.
Чланци 1–18