Home
About Us
Projects
Researchers
All Careers
Internships
Publications
Locations
External Research Office

Ari Kobren

Oracle Labs
Researchers

Ari Kobren

Principal Research Scientist

Ari Kobren

Ari is a research scientist in the machine learning research group (MLRG). Currently, he is generally focused on large language models, and specifically on language agents, decoding algorithms and hallucination. Ari's work is motivated by a desire to ensure that deployed machine learning models are safe, reliable, performant, and easy to use. Ari is also broadly interested in machine learning, algorithmic fairness, and algorithms.

External website: https://akobre01.github.io/

Publications

Slides

EMNLP'22 Presentation of Proxy Clean Work: Mitigating Bias by Proxy in Pre-Trained Models

Swetasudha Panda, Ari Kobren, Michael Wick, Qinlan Shen

Conference Publication

Feeling Validated: Constructing Validation Sets for Few-Shot Intent Classification

Ari Kobren, Naveen Jafer Nizar, Swetasudha Panda, Qinlan Shen, Michael Wick, Jason Peck, Gioacchino Tangari

Conference Publication

Feeling Validated: Constructing Validation Sets for Few-Shot Learning

Ari Kobren, Michael Wick, Swetasudha Panda, Jason Peck, Naveen Jafer Nizar, Qinlan Shen, Gioacchino Tangari

Conference Publication

Upstream Mitigation Is Not All You Need: Testing the Bias Transfer Hypothesis in Pre-Trained Language Models

Ryan Steed, Swetasudha Panda, Michael Wick, Ari Kobren

Slides

Upstream Mitigation Is Not All You Need

Ryan Steed, Michael Wick, Ari Kobren, Swetasudha Panda

Conference Publication

Online Selection with Cumulative Fairness Constraints

Ari Kobren, Swetasudha Panda, Michael Wick

Conference Publication

Diverse Data Augmentation via Unscrambling Text with Missing Words

Ari Kobren, Naveen Jafer Nizar, Michael Wick, Swetasudha Panda

Conference Publication

Searching Near and Far for Examples in Data Augmentation

Ari Kobren, Naveen Jafer Nizar, Michael Wick, Swetasudha Panda

Conference Publication

Searching Near and Far for Examples in Data Augmentation

Ari Kobren, Naveen Jafer Nizar, Michael Wick, Swetasudha Panda

In Proceedings

Online Post-Processing in Rankings for Fair Utility Maximization

Gupta, Ananya*, Johnson, Eric*, Payan, Justin, Roy, Aditya Kumar, Kobren, Ari, Panda, Swetasudha, Tristan, Jean-Baptiste, Michael Wick

In Proceedings

Leveraging Extracted Model Adversaries for Improved Black Box Attacks

Naveen Jafer Nizar, Ari Kobren

In Proceedings

Scalable Hierarchical Clustering with Tree Grafting.

Nicholas Monath, Ari Kobren, Akshay Krishnamurthy, Michael R. Glass, Andrew McCallum

In Proceedings

Offline Contextual Bandits with High Probability Fairness Guarantees.

Blossom Metevier, Stephen Giguere, Sarah Brockman, Ari Kobren, Yuriy Brun, Emma Brunskill, Philip S. Thomas

In Proceedings

Optimal Transport-based Alignment of Learned Character Representations for String Similarity.

Derek Tam, Nicholas Monath, Ari Kobren, Aaron Traylor, Rajarshi Das, Andrew McCallum

In Proceedings

Constructing High Precision Knowledge Bases with Subjective and Factual Attributes.

Ari Kobren, Pablo Barrio 0002, Oksana Yakhnenko, Johann Hibschman, Ian Langmore

In Proceedings

Supervised Hierarchical Clustering with Exponential Linkage.

Nishant Yadav, Ari Kobren, Nicholas Monath, Andrew McCallum

In Proceedings

Paper Matching with Local Fairness Constraints.

Ari Kobren, Barna Saha, Andrew McCallum

In Proceedings

Compact Representation of Uncertainty in Clustering.

Craig Greenberg, Nicholas Monath, Ari Kobren, Patrick Flaherty, Andrew McGregor 0001, Andrew McCallum

In Proceedings

A Hierarchical Algorithm for Extreme Clustering.

Ari Kobren, Nicholas Monath, Akshay Krishnamurthy, Andrew McCallum

In Proceedings

Getting More for Less - Optimized Crowdsourcing with Dynamic Tasks and Goals.

Ari Kobren, Chun How Tan, Panagiotis G. Ipeirotis, Evgeniy Gabrilovich

In Proceedings

Assessing confidence of knowledge base content with an experimental study in entity resolution.

Michael L. Wick, Sameer Singh 0001, Ari Kobren, Andrew McCallum

In Proceedings

Universal Schema for Slot Filling and Cold Start - UMass IESL at TACKBP 2013.

Sameer Singh 0001, Limin Yao, David Belanger, Ari Kobren, Sam Anzaroot, Mike Wick, Alexandre Passos, Harshal Pandya, Jinho D. Choi, Brian Martin, Andrew McCallum

Ari Kobren

Principal Research Scientist

Publications

EMNLP'22 Presentation of Proxy Clean Work: Mitigating Bias by Proxy in Pre-Trained Models

Feeling Validated: Constructing Validation Sets for Few-Shot Intent Classification

Feeling Validated: Constructing Validation Sets for Few-Shot Learning

Upstream Mitigation Is Not All You Need: Testing the Bias Transfer Hypothesis in Pre-Trained Language Models

Upstream Mitigation Is Not All You Need

Online Selection with Cumulative Fairness Constraints

Diverse Data Augmentation via Unscrambling Text with Missing Words

Searching Near and Far for Examples in Data Augmentation

Searching Near and Far for Examples in Data Augmentation

Online Post-Processing in Rankings for Fair Utility Maximization

Leveraging Extracted Model Adversaries for Improved Black Box Attacks

Scalable Hierarchical Clustering with Tree Grafting.

Offline Contextual Bandits with High Probability Fairness Guarantees.

Optimal Transport-based Alignment of Learned Character Representations for String Similarity.

Constructing High Precision Knowledge Bases with Subjective and Factual Attributes.

Supervised Hierarchical Clustering with Exponential Linkage.

Paper Matching with Local Fairness Constraints.

Compact Representation of Uncertainty in Clustering.

A Hierarchical Algorithm for Extreme Clustering.

Getting More for Less - Optimized Crowdsourcing with Dynamic Tasks and Goals.

Assessing confidence of knowledge base content with an experimental study in entity resolution.

Universal Schema for Slot Filling and Cold Start - UMass IESL at TACKBP 2013.

Resources For

Partners

Emerging Technology

What’s New

Contact Us