Imene Kerboua

Imene Kerboua

I’m Imene (pronounced /iman/), a final-year PhD candidate at INSA Lyon and Esker working on NLP and agents for web navigation. Before my PhD, I worked almost two years as a Data Scientist. I love open source: I worked on the French extension of MTEB and on MMTEB, which scales MTEB to many languages, especially low-resource ones. Outside research, I enjoy hiking, reading books and painting.

Projects

FocusAgent

Research project

Simple yet effective ways of trimming the large context of web agents: a small retriever model reads the page and keeps only the lines that matter, so the acting model sees a much smaller observation — and never reads pruned prompt-injection content.

  • Web agents
  • LLM
  • Prompt injection
Visit

MMTEB

Contributor

Scaling MTEB to many languages, with a focus on low-resource ones.

  • Benchmarking
  • Embeddings
  • Open source
Visit

MTEB French

Contributor

Extension of the Massive Text Embedding Benchmark to French.

  • NLP
  • French
  • Open source
Visit

Publications

  1. Does The Way You Plan Matter? An Empirical Study of Planning Representations for LLM Web Agents

    Alejandra Zambrano, Sara Vera Marjanovic, Imene Kerboua, Xing Han Lù, Leila Kosseim

    arXiv preprint · 2026

  2. CUBE: A Standard for Unifying Agent Benchmarks

    Alexandre Lacoste, Nicolas Gontier, Oleh Shliazhko, …, Imene Kerboua, …, Siva Reddy, Dawn Song

    arXiv preprint · 2026

  3. WebArena-Pro: A Heterogeneous, Multimodal, Reproducible Benchmark for Web Agents

    Imene Kerboua, Fatemeh Pesaran Zadeh, Xing Han Lù, et al.

    Workshop on Agents in the Wild: Safety, Security, and Beyond · 2026

  4. FocusAgent: Simple Yet Effective Ways of Trimming the Large Context of Web Agents

    Imene Kerboua, Sahar Omidi Shayegan, Megh Thakkar, Xing Han Lù, Léo Boisvert, Massimo Caccia, Jérémy Espinas, Alexandre Aussem, Véronique Eglin, Alexandre Lacoste

    arXiv preprint · 2025

  5. LineRetriever: Planning-Aware Observation Reduction for Web Agents

    Imene Kerboua, Sahar Omidi Shayegan, Megh Thakkar, Xing Han Lù, Massimo Caccia, Véronique Eglin, Alexandre Aussem, Jérémy Espinas, Alexandre Lacoste

    arXiv preprint · 2025

  6. Maintaining MTEB: Towards Long Term Usability and Reproducibility of Embedding Benchmarks

    Isaac Chung, Imene Kerboua, Márton Kardos, Roman Solomatin, Kenneth Enevoldsen

    arXiv preprint · 2025

  7. MIEB: Massive Image Embedding Benchmark

    Chenghao Xiao, Isaac Chung, Imene Kerboua, Jamie Stirling, Xin Zhang, Márton Kardos, Roman Solomatin, Noura Al Moubayed, Kenneth Enevoldsen, Niklas Muennighoff

    ICCV 2025 · 2025

  8. MMTEB: Massive Multilingual Text Embedding Benchmark

    Kenneth Enevoldsen, Isaac Chung, Imene Kerboua, Márton Kardos, Ashwin Mathur, David Stap, …, Niklas Muennighoff

    ICLR 2025 · 2025

  9. MTEB-FR : une expérience à large échelle pour l'apprentissage de représentation en français

    Wissam Siblini, Mathieu Ciancone, Imene Kerboua, Marion Schaeffer

    EGC 2025 · 2025

  10. MTEB-French: Resources for French Sentence Embedding Evaluation and Analysis

    Mathieu Ciancone, Imene Kerboua, Marion Schaeffer, Wissam Siblini

    arXiv preprint · 2024

Teaching

Guest Lecture on LLM Agents

Guest lecture ·McGill University ·2025

Montréal, Canada

Invited two-hour guest lecture on LLM agents at McGill University.

AI Agents

CS Master's course ·aivancity ·2025

Paris, France

Introduction to AI Agents, specificaly to LLM agents.

Cloud Computing

CS Master's course ·Université Lumière, Lyon 2, Computer Science Department ·2024

Lyon, France

Introduction to Cloud Computing using Azure.

Object Oriented Programming with Python

Master's course ·Groupe GEMA ·2022 – 2023

Lyon, France

Object Oriented Programming with Python for Master’s students: classes, inheritance, and writing maintainable Python.

Advanced SQL

Master's course ·Groupe GEMA ·2022 – 2023

Lyon, France

Advanced SQL for Master’s students: complex queries, joins, window functions and query optimisation.

Resume

Work experience

  • AI PhD Student·Esker

    May 2023 – present

    • CIFRE partnership with LIRIS and INSA Lyon.
    • Working on autonomous web agents and embeddings evaluation.

  • Teacher Assistant·Université Lumière Lyon 2

    September 2024 – February 2026

    • Cloud Computing with Azure (Master's).
    • MLOps with Azure (Master's).

  • Visiting Researcher·Mila – Quebec AI Institute

    August 2025 – November 2025

    • Research visit in Montréal, Canada.

  • Lecturer·Groupe GEMA

    October 2022 – January 2023

    • Part-time Master's classes, alongside the Data Scientist role at Cygogn.
    • Object Oriented Programming with Python, advanced SQL and NLP.

  • Data Scientist·Cygogn

    September 2021 – April 2023

    • Model development and deployment on AWS.
    • BI and data engineering.

  • Data Scientist NLP·Once For All France

    March 2021 – September 2021

    • Developed models for email classification.

  • Frontend Web Developer·SFR Business

    June 2020 – September 2020

    • Developed an app for data visualization using Symfony and ReactJS.

  • Machine Learning Researcher·LRIA Lab, USTHB

    February 2019 – June 2019

    • Conception and implementation of a hybrid recommender system based on deep learning.

Education

  • Ph.D in Machine Learning·INSA Lyon

    December 2023 – December 2026

  • M.S. in Computer Science, Data Mining·Université Lumière Lyon 2

    September 2019 – September 2021

  • B.S. in Information Systems and Software Engineering·USTHB

    September 2016 – July 2019

Skills

  • LLMs
  • Agents
  • RAG
  • Azure
  • Python
  • Git
  • Docker
  • SQL
  • AWS
  • Machine Learning
  • Deep Learning
  • NLP