Weekly Media Update

Week of July 18, 2025

Warren Center Affiliates at ICML 2025 

Twenty-six papers, posters, workshops, and tutorials from Warren Center faculty affiliates (including three Spotlight Posters) have been accepted to the Forty-Second International Conference on Machine Learning (ICML 2025). A huge congratulations to these affiliates for their impressive research. See more about their work in the links below and in next Friday's email.

Jonathan Moreno on Bioethics and the Rules-Based International Order

Attorney and bioethicist Karolina Zhukoff sat down with Jonathan Moreno to discuss his new book, “Absolutely Essential: Bioethics and the Rules-Based International Order.”  They delved into the ways in which Moreno’s upbringing inspired his work, the global impact of COVID-19 on bioethics, and the challenges posed to the international rules-based order by geopolitical shifts and emerging technologies. 

Alignment of Large Language Models With Constrained Learning

7/18: MoFA at ICML 2025 | Alejandro Ribeiro 

Composition and Alignment of Diffusion Models Using Constrained Learning

7/18: MoFA at ICML 2025 | Alejandro Ribeiro 

Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agents

7/17: ICML 2025 Spotlight Poster | Aaron Roth and Hamed Hassani

Gradient Flow Provably Learns Robust Classifiers for Orthonormal GMMs

7/17: ICML 2025 Poster | René Vidal

GIVE: Structured Reasoning of Large Language Models With Knowledge Graph Inspired Veracity Extrapolation

7/17: ICML 2025 Poster | Alejandro Ribeiro

Sum-of-Parts: Self-Attributing Neural Networks With End-to-End Learning of Feature Groups

7/17: ICML 2025 Poster | Bhuvnesh Jain

CoMemo: LVLMs Need Image Context With Image Memory

7/17: ICML 2025 Poster | Weijie Su 

Adversarial Reasoning at Jailbreaking Time

7/16: ICML 2025 Poster | Hamed Hassani

Guarantees of a Preconditioned Subgradient Algorithm for Overparameterized Asymmetric Low-Rank Matrix Recovery

7/16: ICML 2025 Poster | René Vidal

Reconstructing Cell Lineage Trees From Phenotypic Features With Metric Learning

7/16: ICML 2025 Poster | Junhyong Kim

Stronger Neyman Regret Guarantees for Adaptive Experimental Design

7/16: ICML 2025 Spotlight Poster | Aaron Roth

High-Dimensional Prediction for Sequential Decision Making

7/16: ICML 2025 Spotlight Poster and Oral Presentation | Aaron Roth

The Relationship Between No-Regret Learning and Online Conformal Prediction

7/16: ICML 2025 Poster | Aaron Roth

Intersectional Fairness in Reinforcement Learning With Large State and Constraint Spaces

7/15: ICML 2025 Poster | Michael Kearns and Aaron Roth

Stochastic Online Conformal Prediction With Semi-Bandit Feedback

7/15: ICML 2025 Poster | Hamsa Bastani

On The Concurrence of Layer-Wise Preconditioning Methods and Provable Feature Learning

7/15: ICML 2025 Poster | Hamed Hassani and Nik Matni

Restoring Calibration for Aligned Large Language Models: A Calibration-Aware Fine-Tuning Approach

7/15: ICML 2025 Poster | Weijie Su 

A Manifold Perspective on the Statistical Generalization of Graph Neural Networks

7/15: ICML 2025 Poster | Alejandro Ribeiro

The Road to Autonomous Trucking: Scale, Feasibility, and Economic Impact

7/14: Gad's Newsletter | Gad Allon 

Judges Block Trump Agenda Despite SCOTUS

7/14: Bloomberg Law Podcast | Cary Coglianese 

Jailbreaking LLMs and Agentic Systems: Attacks, Defenses, and Evaluations

7/14: ICML 2025 Tutorial | Hamed Hassani

The Summer Institute in Computational Social Science

7/11: University of Pennsylvania | Dani Bassett and Lyle Ungar

Voice-Based Chatbots for English Speaking Practice in Multilingual Low-Resource Schools: A Multi-Stakeholder Mixed-Methods Study

7/11: VeriXiv | Lyle Ungar

Artificial Intelligence Could Supercharge Trump’s Deregulatory Push, but Experts Flag Shortfalls

7/10: Government Executive | Cary Coglianese

Auditors Can’t Save Carbon Offsets

7/10: Science | Cary Coglianese 

Auditors Fail in Role of Safeguarding Carbon Offsets: Study

7/10: Bloomberg | Cary Coglianese 

Trump Agenda Being Blocked by Judges Despite Supreme Court

7/10: Bloomberg | Cary Coglianese 

Fast-Casual Restaurants Lean on Loyalty Programs to Offset Consumer Pullback

7/5: CNBC | Peter Fader

Voyaging Into Unbounded Dynamic Scenes From a Single View

7/5: arXiv | René Vidal

Conformal Information Pursuit for Interactively Guiding Large Language Models

7/4: arXiv | Edgar Dobriban, Hamed Hassani, and René Vidal

Reply to Dablander et al.: Identifying Interventions That Change Intentions Lays a Valuable Foundation for Behavior Change

7/3: PNAS | Emily Falk

Website

Get Involved

Past Events

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