cv

You can download my CV in PDF format here.

Basics

Name Ahmad Rahimi
Label Ph.D. Candidate
Email ahmad.rahimi@epfl.ch
Phone +41 76 266 94 37
Url https://AhmadRHM.github.io/
Summary PhD candidate at EPFL advised by Prof. Alexandre Alahi, specializing in efficient video generation, driving world models, and 3D/4D scene reconstruction. I will complete my PhD in February 2027 and am interested in research scientist and research engineer opportunities beginning in March 2027.

Interests

Deep Learning
World Modeling
Video Generation
Trajectory Prediction
Autonomous Driving
Computer Vision

Education

  • 2022.09 - Present

    Lausanne, Switzerland

    PhD
    Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
    Computer Science
  • 2018.09 - 2022.06

    Tehran, Iran

    BSc
    Sharif University of Technology, Tehran, Iran
    Computer Science

Work

  • 2025.11 - 2026.05
    Applied Science Intern
    Wayve
    Worked on world modeling, VR-based driving-scene analysis, and 4D scene reconstruction. I developed a VR tool for interactively analyzing driving scenes and methods for reconstructing and iteratively refining dynamic driving scenes using Gaussian splatting.

Publications

  • 2025.12.01
    MAD: Motion Appearance Decoupling for efficient Driving World Models
    CVPR 2026
    We propose a two-step approach to driving world models that separates generation into motion forecasting and appearance synthesis. We fine-tune off-the-shelf video generation models into two components: a motion forecaster that predicts future object movements in an abstract skeleton representation, and an appearance synthesizer that converts this motion into realistic video. Our approach is effective and efficient, requiring only 140 GPU hours and driving videos for training, being two orders of magnitude less than prior work.
  • 2025.06.08
    From Generation to Generalization: Emergent Few-Shot Learning in Video Diffusion Models
    ICML 2026
    We introduce a few-shot fine-tuning framework that repurposes pretrained video diffusion models for novel visual and logic tasks using only a handful of examples. The updated version of this work, Rethinking Visual Intelligence: Insights from Video Pretraining, received an ARC Prize 2025 Honorable Mention.
  • 2024.12.03
    GEM: A Generalizable Ego-vision Multimodal World Model for Fine-Grained Ego-Motion, Object Dynamics, and Scene Composition Control
    CVPR 2025
    We propose a generalizable ego-vision multimodal world model, GEM, that predicts future frames of egocentric driving videos using multiple control signals. We build on a pretrained Stable Video Diffusion model, adding new components for fine-grained controllability and long-term generation.
  • 2024.12.02
    A Multi-Loss Strategy for Vehicle Trajectory Prediction: Combining Off-Road, Diversity, and Directional Consistency Losses
    Under review
    Traditionally, trajectory prediction models are trained using minADE losses, which only penalize the closest prediction to the ground truth. This sparsification of the loss function slows down model convergence and may lead to suboptimal predictions for less trained prediction heads. In this work, we propose a multi-loss strategy that combines off-road, diversity, and directional consistency losses which act on all prediction modes and improve the performance of trajectory prediction models.
  • 2023.12.07
    Sim-to-Real Causal Transfer: A Metric Learning Approach to Causally-Aware Interaction Representations
    CVPR 2025
    We investigate causal understanding in the context of multi agent interaction prediction, where removing non-causal agents from the scene should not change model's prediction. We show modern prediction models are able to identify non-causal agents in the scene, but fail to properly model causal agent removals. We propose a metric learning approach to learn causally-aware interaction representations, which we show lead to better generalizability and robustness for Out of Distribution (OOD) and low-data regime scenarios.
  • 2022.03.29
    Vehicle trajectory prediction works, but not everywhere
    CVPR 2022
    We show that current trajectory prediction models fail to fully understand scene structure, where naturalistic perturbations in the scene, like introducing turns, can significantly affect the prediction quality of the model. We propose a scene attack method to evaluate the robustness of trajectory prediction models against such perturbations, and further show improvements of the models when trained with these adversarial examples.

Awards

  • 2025.12.05
    ARC Prize 2025 Honorable Mention
    ARC Prize Foundation
    Recognized for research on repurposing video diffusion models for novel visual and logic tasks.
  • 2017.09.01
    Bronze Medal in INOI
    Iranian National Olympiad in Informatics
    Awarded the Bronze Medal in the Iranian National Olympiad in Informatics (INOI), which evaluates theoretical and practical skills in algorithms and combinatorics.
  • 2021.09.01
    Ranked Second in !Optimizer Competition
    Sharif University of Technology
    Ranked second in the !Optimizer competition, a competition held by the computer science department at Sharif University of Technology, which evaluates students' knowledge in optimization.
  • 2018.09.01
    Ranked 403 in Nation-wide university entrance exam
    Iranian National Evaluation Organization
    Ranked 403 in the nation-wide university entrance exam, among more than 300,000 participants in Iran (placed in top 0.2% of the participants).
  • 2019.12.07
    Ranked Third in Developers Competition
    Sharif University of Technology
    Ranked third in the developers competition, a competition held by the computer engineering department at Sharif University of Technology, which evaluates students' knowledge in web development.

Volunteer

  • 2023.08 - 2024.08

    Lausanne, Switzerland

    President
    Iranian Student Association (IRSA)
    President of Iranian Student Association at EPFL, where I organized various cultural events and workshops for Iranian students at EPFL.
  • 2020.01 - 2020.01

    Tehran, Iran

    Organizer
    Code Knock 3
    Organizer of Code Knock 3, a competitive programming competition in Java held at Sharif University of Technology.
  • 2019.08 - 2019.08

    Tehran, Iran

    Lead Organizer
    Sharif Math Summer School
    Lead organizer of the Sharif Math Summer School, a summer school for high school students to familiarize them with various mathematical fields. I was also the lead of the cryptography workshop.
  • 2019.06 - 2020.07

    Tehran, Iran

    President
    Student Scientific Association
    President of the Student Scientific Association in Department of Mathematical Sciences at Sharif University of Technology, where I organized various scientific events and workshops for students.

Skills

Machine Learning/Deep Learning
PyTorch
TensorFlow
Keras
Scikit-learn
OpenCV
Programming
Python
C++
Java
Tools
Git
Docker
Jupyter
PyCharm

Languages

Persian
Native speaker
English
Fluent
French
Intermediate

References

Professor Alexandre Alahi
He is my current PhD advisor at EPFL. He is an expert in computer vision and deep learning for autonomous driving, and has published numerous papers in top-tier conferences.
Seyed Mohsen Moosavi-Dezfooli
He was a supervisor for an internship I did at EPFL and we have published a paper together. He is an expert in adversarial machine learning and has published numerous papers in top-tier conferences.
Hugh Tomkins
He was my supervisor during my Applied Science internship at Wayve, where I worked on VR-related projects and 4D Gaussian splatting for scene reconstruction and generation.