SIMA 2: A Generalist Embodied Agent for Virtual Worlds

SIMA 2 Team

DeepMind Release, Dec 2025

Paper / Blog

A Clean Slate for Offline Reinforcement Learning

MT Jackson*, U Berdica*, J Liesen*, S Whiteson, JN Foerster

NeurIPS 2025 (Oral)

Paper / Code / Talk

Token-Sparse Diffusion Transformers

MT Jackson, B Ellis, S Whiteson, JN Foerster

Under review

Paper

Jafar: An Open-Source Genie Reimplemention in Jax

T Willi*, MT Jackson*, JN Foerster

ICML 2024 Workshop on Controllable Video Generation

Code

Policy-Guided Diffusion

MT Jackson, MT Matthews, C Lu, B Ellis, J Foerster, S Whiteson

RLC 2024 - NeurIPS 2023 Workshop on Robot Learning

Paper / Code

Adam on Local Time: Addressing Nonstationarity in RL with Relative Adam Timesteps

B Ellis*, MT Jackson*, A Lupu, AD Goldie, M Fellows, S Whiteson, JN Foerster

NeurIPS 2024

Paper

Discovering Temporally-Aware Reinforcement Learning Algorithms

MT Jackson*, C Lu*, L Kirsch, RT Lange, S Whiteson, JN Foerster

ICLR 2024

Paper / Code / Podcast

Discovering General Reinforcement Learning Algorithms with Adversarial Environment Design

MT Jackson, M Jiang, J Parker-Holder, R Vuorio, C Lu, G Farquhar, S Whiteson, JN Foerster

NeurIPS 2023

Paper / Code

Imagined Autocurricula

AH Guzel, MT Jackson, JL Liesen, T Rocktaschel, JN Foerster, I Bogunovic, J Parker-Holder

NeurIPS 2025

Paper

An Optimisation Framework for Unsupervised Environment Design

N Monette, A Letcher, M Beukman, MT Jackson, A Rutherford, AD Goldie, JN Foerster

RLC 2025

Paper

Judge a Book by Its Cover: Investigating Multi-Modal LLMs for Multi-Page Handwritten Document Transcription

B Gutteridge, MT Jackson, T Kukurin, X Dong

AAAI 2025 DocUI Workshop

Paper

Can Learned Optimization Make Reinforcement Learning Less Difficult?

AD Goldie, C Lu, MT Jackson, S Whiteson, JN Foerster

NeurIPS 2024 (Spotlight)

Paper

Craftax: A Lightning-Fast Benchmark for Open-Ended Reinforcement Learning

M Matthews, M Beukman, B Ellis, M Samvelyan, MT Jackson, S Coward, J Foerster

ICML 2024 (Spotlight)

Paper / Blog / Code

Risks and Opportunities of Open Source Generative AI

F Eiras, A Petrov, B Vidgen, C Schroeder de Witt, F Pizzati, K Elkins, S Mukhopadhyay, A Bibi, B Csaba, F Steibel, F Barez, G Smith, G Guadagni, J Chun, J Cabot, JM Imperial, JA Nolazco-Flores, L Landay, MT Jackson, P Röttger, PHS Torr, T Darrell, YS Lee, J Foerster

ICML 2024 (Oral)

Paper

SplAgger: Split Aggregation for Meta-Reinforcement Learning

J Beck, MT Jackson, R Vuorio, Z Xiong, S Whiteson

RLC 2024

Paper

Retrieve What You Need: A Mutual Learning Framework for Open-domain Question Answering

D Wang, Q Huang, MT Jackson, J Gao

TACL 2024

Paper

Online Reinforcement Learning Controllers for Soft Robots using Learned Environments

U Berdica, MT Jackson, J Foerster, P Maiolino

RoboSoft 2024

Hypernetworks in Meta-Reinforcement Learning

J Beck, MT Jackson, R Vuorio, S Whiteson

CoRL 2022

Paper

Multi-Modal Fusion by Meta-Initialization

MT Jackson*, SA Malik*, MT Matthews, Y Mohamed-Ahmed

FARSCOPE Robotics Workshop 2022 (Best Poster Award)

Paper