Weekly Media Update

Week of December 13, 2024

NeurIPS 2024 Papers from Warren Center Faculty Affiliates

Fifteen papers from Warren Center faculty affiliates, including three Spotlight papers, have been accepted to the Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS 2024). A huge congratulations to these affiliates for their impressive research. See more about the papers in the links below.

Penn Professor Wins Prestigious Award for Drone Research

Drones continue to evolve as an important way to aid and support humans. In this interview on PHL17 Morning News, Vijay Kumar discusses his research to enhance a drone’s reach and his recent John Scott Award. "We do things that benefit humankind," says Kumar. 

Why Income Inequality in America Is Hurting Collective Happiness

What does income inequality in America have to do with happiness? And more generally, what is the right way to think about the relationship between money and happiness? A recent study by Matthew Killingsworth offers some surprising answers.

One-Shot Safety Alignment for Large Language Models via Optimal Dualization

12/13: NeurIPS 2024 Spotlight poster | Edgar Dobriban and Hamed Hassani

JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models

12/13: NeurIPS 2024 poster | Edgar Dobriban and Hamed Hassani

The Evolution of Statistical Induction Heads: In-Context Learning Markov Chains

12/13: NeurIPS 2024 poster | Surbhi Goel

Neural decoding from stereotactic EEG: accounting for electrode variability across subjects

12/13: NeurIPS 2024 poster | Konrad Kording

Geometric Analysis of Nonlinear Manifold Clustering

12/13: NeurIPS 2024 poster | René Vidal

Generative Adversarial Model-Based Optimization via Source Critic Regularization

12/12: NeurIPS 2024 poster | Hamsa Bastani 

Tolerant Algorithms for Learning with Arbitrary Covariate Shift

12/12: NeurIPS 2024 Spotlight poster | Surbhi Goel

Length Optimization in Conformal Prediction

12/12: NeurIPS 2024 poster | Hamed Hassani

Reconstruction Attacks on Machine Unlearning: Simple Models are Vulnerable

12/12: NeurIPS 2024 poster | Michael Kearns and Aaron Roth 

Oracle-Efficient Reinforcement Learning for Max Value Ensembles

12/12: NeurIPS 2024 poster | Michael Kearns and Aaron Roth 

Managing the Humans in the AI Loop with Wendy Gonzalez

12/12: The Road to Accountable AI | Kevin Werbach

A Textbook Remedy for Domain Shifts: Knowledge Priors for Medical Image Analysis

12/11: NeurIPS 2024 Spotlight poster | Chris Callison-Burch

PaCE: Parsimonious Concept Engineering for Large Language Models

12/11: NeurIPS 2024 poster | Chris Callison-Burch and René Vidal 

Artificial Intelligence; Regulating the Evolution

12/11: Erie News Now | Cary Coglianese 

Toward Efficient Inference for Mixture of Experts

12/11: NeurIPS 2024 poster | Benjamin Lee

Constrained Diffusion Models via Dual Training

12/11: NeurIPS 2024 poster | Alejandro Ribeiro

Vision Model Pre-training on Interleaved Image-Text Data via Latent Compression Learning

12/11: NeurIPS 2024 poster | Weijie Su

Is This the End of Low-Cost Airlines in the U.S.?

12/9: Gad's Newsletter | Gad Allon

Elon Musk says DOGE can 'gut the federal government' with a recent Supreme Court ruling. Some lawyers disagree.

12/7: Business Insider | Cary Coglianese

AI in the Courts: How Worried Should We Be?

12/5: Judicature | Cary Coglianese

Solutions for an inclusive connected future: Success factors and challenges

12/3: FFD4 | Christopher Yoo

Tesla investors, analysts have questions on the potential of robotaxis and self-driving

12/2: Yahoo Finance | Cary Coglianese

Study design features increase replicability in brain-wide association studies

11/27: Nature | Ted Satterthwaite

People judge others more harshly after talking to bots

9/19: PNAS Nexus | Lyle Ungar

Website

Get Involved

Past Events

107 Towne Building
220 South 33rd Street
Philadelphia, Pennsylvania 19104
warrenct@seas.upenn.edu