papers

Publications (8)

cs.LG2026

Decoupling Exploration and Policy Optimization: Uncertainty Guided Tree Search for Hard Exploration

Zakaria Mhammedi, James Cohan

The process of discovery requires active exploration -- the act of collecting new and informative data. However, efficient autonomous exploration remains a major unsolved problem.…

cs.CY2025

Evaluating Gemini in an arena for learning

LearnLM Team, Abhinit Modi, Aditya Srikanth Veerubhotla +34

Artificial intelligence (AI) is poised to transform education, but the research community lacks a robust, general benchmark to evaluate AI models for learning. To assess state-of-t…

cs.LG2026

Bridging Kolmogorov Complexity and Deep Learning: Asymptotically Optimal Description Length Objectives for Transformers

Peter Shaw, James Cohan, Jacob Eisenstein +1

The Minimum Description Length (MDL) principle offers a formal framework for applying Occam's razor in machine learning. However, its application to neural networks such as Transfo…

cs.LG2025

ALTA: Compiler-Based Analysis of Transformers

Peter Shaw, James Cohan, Jacob Eisenstein +3

We propose a new programming language called ALTA and a compiler that can map ALTA programs to Transformer weights. ALTA is inspired by RASP, a language proposed by Weiss et al. (2…

cs.CL2025

Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

Gheorghe Comanici, Eric Bieber, Mike Schaekermann +3431

In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our…

cs.CY2025

Towards Responsible Development of Generative AI for Education: An Evaluation-Driven Approach

Irina Jurenka, Markus Kunesch, Kevin R. McKee +71

A major challenge facing the world is the provision of equitable and universal access to quality education. Recent advances in generative AI (gen AI) have created excitement about…

cs.LG2023

From Pixels to UI Actions: Learning to Follow Instructions via Graphical User Interfaces

Peter Shaw, Mandar Joshi, James Cohan +6

Much of the previous work towards digital agents for graphical user interfaces (GUIs) has relied on text-based representations (derived from HTML or other structured data sources),…

cs.CY2025

LearnLM: Improving Gemini for Learning

LearnLM Team, Abhinit Modi, Aditya Srikanth Veerubhotla +43

Today's generative AI systems are tuned to present information by default, rather than engage users in service of learning as a human tutor would. To address the wide range of pote…