Yonathan Efroni

Computer Science · Physical Sciences

12h-index381citations65works2.22yr avg
OpenAlex↗Google Scholar↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Reinforcement Learning in Robotics(40), Advanced Bandit Algorithms Research(30), Machine Learning and Algorithms(14), Data Stream Mining Techniques(8), Evolutionary Algorithms and Applications(7)

Publications65 total

Structure Enables Effective Self-Localization of Errors in LLMs
arXiv (Cornell University)·2026Open Access
Structure Enables Effective Self-Localization of Errors in LLMs
ArXiv.org·2026Open Access
Imbalanced Gradients in RL Post-Training of Multi-Task LLMs
arXiv (Cornell University)·2025Open Access
Self-Improvement of Language Models by Post-Training on Multi-Agent Debate
arXiv (Cornell University)·2025Open Access
Gradient Free Deep Reinforcement Learning With TabPFN
arXiv (Cornell University)·2025Open Access
Simple Optimizers for Convex Aligned Multi-Objective Optimization
arXiv (Cornell University)·2025Open Access
Aligned Multi Objective Optimization
arXiv (Cornell University)·2025Open Access
Exploiting Structure in Offline Multi-Agent RL: The Benefits of Low Interaction Rank
arXiv (Cornell University)·2024Open Access
RL in Latent MDPs is Tractable: Online Guarantees via Off-Policy Evaluation
arXiv (Cornell University)·2024Open Access
Generalizing Multi-Step Inverse Models for Representation Learning to Finite-Memory POMDPs
arXiv (Cornell University)·2024Open Access
The Bias of Harmful Label Associations in Vision-Language Models
arXiv (Cornell University)·2024Open Access
Abstraction-Based Decision Making for Statistical Properties (Invited Talk)
DROPS (Schloss Dagstuhl – Leibniz Center for Informatics)·2024· 29 citedOpen Access
RL in Latent MDPs is Tractable: Online Guarantees via Off-Policy Evaluation
2024
Pearl: A Production-ready Reinforcement Learning Agent
arXiv (Cornell University)·2023· 2 citedOpen Access
PcLast: Discovering Plannable Continuous Latent States
arXiv (Cornell University)·2023Open Access
Prospective Side Information for Latent MDPs
arXiv (Cornell University)·2023Open Access
Agent-Controller Representations: Principled Offline RL with Rich Exogenous Information
arXiv (Cornell University)·2022Open Access
Reward-Mixing MDPs with a Few Latent Contexts are Learnable
arXiv (Cornell University)·2022Open Access
Tractable Optimality in Episodic Latent MABs
arXiv (Cornell University)·2022Open Access
Guaranteed Discovery of Control-Endogenous Latent States with Multi-Step Inverse Models
arXiv (Cornell University)·2022· 1 citedOpen Access
page 1 of 4Next →

Frequent Co-authors

Shie Mannor(34), Mohammad Ghavamzadeh(9), Gal Dalal(8), Bruno Scherrer(8), Lior Shani(6), Nadav Merlis(6), Jeongyeol Kwon(6), Constantine Caramanis(5), Manan Tomar(4), Aviv Rosenberg(3), Guy Tennenholtz(3), Uri Shalit(3), Alex Lamb(3), Riashat Islam(3), Akshay Krishnamurthy(3), John Langford(3), Ayush Jain(3), Kaveh Hassani(3), Boris Vidolov(3), Chen Tessler(2)