Publications by Zico Kolter

Preprint

A New Framework for Cybersecurity Refusals in AI Agents

2026
Jones EK, Dziemian M, Fredrikson M, Kolter JZ

Preprint

Accelerating Diffusion Planners in Offline RL via Reward-Aware Consistency Trajectory Distillation

2026
Duan X, He Y, Tajwar F, Salakhutdinov R, Kolter JZ, Schneider J

Preprint

Antidistillation Fingerprinting

2026
Xu YE, Kirchenbauer J, Savani Y, Trockman A, Robey A, Goldstein T, Fang F, Kolter JZ

Preprint

Base Models Look Human To AI Detectors

2026
Xu YE, Zhong Z, Raghunathan A, Fang F, Kolter JZ

Journal Article

Beyond alignment: Why robotic foundation models need context-aware safety

2026 • Science Robotics • 11(113):
Robey A, Ravichandran Z, Jones EK, Perlo J, Barez F, Kumar V, Kolter JZ, Hassani H, Pappas GJ

Preprint

Comparing AI Agents to Cybersecurity Professionals in Real-World Penetration Testing

2026
Lin JW, Jones EK, Jasper DJ, Ho EJ-S, Wu A, Yang AT, Perry N, Zou A, Fredrikson M, Kolter JZ, Liang P, Boneh D, Ho DE

Preprint

Compressed Sensing for Capability Localization in Large Language Models

2026
Bair A, Xu YE, Sun M, Kolter JZ

Preprint

From Entropy to Epiplexity: Rethinking Information for Computationally Bounded Intelligence

2026
Finzi M, Qiu S, Jiang Y, Izmailov P, Kolter JZ, Wilson AG

Preprint

Improved Mean Flows: On the Challenges of Fastforward Generative Models

2026
Geng Z, Lu Y, Wu Z, Shechtman E, Kolter JZ, He K

Preprint

Joint Distillation for Fast Likelihood Evaluation and Sampling in Flow-based Models

2026
Ai X, He Y, Gu A, Salakhutdinov R, Kolter JZ, Boffi NM, Simchowitz M

Preprint

Mamba-3: Improved Sequence Modeling using State Space Principles

2026
Lahoti A, Li KY, Chen B, Wang C, Bick A, Kolter JZ, Dao T, Gu A

Preprint

Measuring Five-Nines Reliability: Sample-Efficient LLM Evaluation in Saturated Benchmarks

2026
Kim E, Gu C, Tiwari V, Kolter JZ

Preprint

Mimetic Initialization of MLPs

2026
Trockman A, Kolter JZ

Preprint

Not All Rollouts are Useful: Down-Sampling Rollouts in LLM Reinforcement Learning

2026
Xu YE, Savani Y, Fang F, Kolter JZ

Preprint

Reevaluating Policy Gradient Methods for Imperfect-Information Games

2026
Rudolph M, Lichtle N, Mohammadpour S, Bayen A, Kolter JZ, Zhang A, Farina G, Vinitsky E, Sokota S

Preprint

ROGUE: Evaluating Corrigibility Failures in Frontier Computer-Use Agents

2026
Tien J, Anand A, Tuan Y-R, Shen Y, Kolter JZ, Nayebi A

Preprint

Scalable Decision Making for Games of Imperfect Information

2026
Sokota S, Vinitsky E, Hu H, Fan Z, Kolter JZ, Farina G

Journal Article

Scalable decision-making for games of imperfect information

2026 • Nature • 658(8134):
Sokota S, Vinitsky E, Hu H, Fan Z, Kolter JZ, Farina G

Preprint

Sharding Prevents LLM Oversight Failures and Adversarial Exploitation

2026
Akinwande V, Kolter JZ, Nayebi A

Preprint

The Finetuner's Fallacy: When to Pretrain with Your Finetuning Data

2026
Kolter Z, and others

Preprint

The Mixing method: low-rank coordinate descent for semidefinite programming with diagonal constraints

2026
Wang P-W, Chang W-C, Kolter JZ

Preprint

Understanding and Mitigating Premature Confidence for Better LLM Reasoning

2026
Gai J, Zeng G, Baek C, Wu C, Kolter JZ, Risteski A, Raghunathan A

Preprint

When Should We Introduce Safety Interventions During Pretraining?

2026
Sam D, Goyal S, Maini P, Robey A, Kolter JZ

Conference

ADAPTIVE DATA OPTIMIZATION: DYNAMIC SAMPLE SELECTION WITH SCALING LAWS

2025 • 13th International Conference on Learning Representations Iclr 2025 • 19718-19741
Jiang Y, Zhou A, Feng Z, Malladi S, Kolter JZ

Preprint

Adversarial Attacks on Robotic Vision Language Action Models

2025
Jones EK, Robey A, Zou A, Ravichandran Z, Pappas GJ, Hassani H, Fredrikson M, Kolter JZ

At a Glance

Academic Offerings

Admissions

Directory Submenu

People

Explore the Field

Publications by Zico Kolter

Preprint

Preprint

Preprint

Preprint

Journal Article

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Journal Article

Preprint

Preprint

Preprint

Preprint

Preprint

Conference

Preprint

At a Glance

Academic Offerings

Admissions

Directory Submenu

People

Explore the Field

What can we help you find?

Preprint

Preprint

Preprint

Preprint

Journal Article

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Preprint

Journal Article

Preprint

Preprint

Preprint

Preprint

Preprint

Conference

Preprint