FullTime with BenefitsLog in

Master Internship - Tool‐augmented LLM agents for abdominal CT question answering: benchmarking, tool attribution and reasoning faithfulness

Inria · Gif-sur-Yvette

From inria.fr · seen open 19 November 2026 · checked before listing

FWB read this posting on inria.fr on 19 November 2026 and looks every day whether it is still open. Before it was listed, it was checked for the signs of a scam. Something wrong? Report it

  • Internship
  • Posted on 19 November 2026
  • Not on FWB yet · apply on its site
  • LLM
  • LLM agents
  • AI agents
  • CT
  • Benchmarking
  • Attribution

Apply on inria.fr (opens in a new tab)Save on FWB

Contenu principal

Accessibilité

Contact

Mentions légales

Menu

Fermer

FR EN

Se connecter

Offres d'emploi Inria >

Offre n°2026-10733

Master Internship - Tool‐augmented LLM agents for abdominal CT question answering: benchmarking, tool attribution and reasoning faithfulness

Télécharger l'offre au format PDF

Le descriptif de l’offre ci-dessous est en Anglais

Type de contrat : Stage

Niveau de diplôme exigé : Bac + 5 ou équivalent

Fonction : Stagiaire de la recherche

Niveau d'expérience souhaité : Jeune diplômé

A propos du centre ou de la direction fonctionnelle

Created in 2008, the Inria Saclay Center is located at the heart of the Paris-Saclay scientific and technological excellence cluster, which alone accounts for 15% of French research. Serving the development of the Université Paris-Saclay and the Institut Polytechnique de Paris, the Inria Saclay center employs 80 people in research support services and 500 scientists of 54 nationalities.

Benefiting from continuous growth, the center now has a total of 42 project-teams and two in the process of being created, including 21 jointly with the Institut Polytechnique de Paris, 16 with the Université Paris-Saclay, as well as 7 Inria EPs, including one in collaboration with Onera and one with the Pôle Universitaire Centre Val de Loire. These research teams are spread over more than ten sites.

Contexte et atouts du poste

In the context of the ANR FM2AI, and in collaboration with APHP, the aim of this internship is to systematically evaluate LLM-based agents that combine reasoning with specialized medical imaging tools for abdominal CT question answering, focusing on their accuracy, computational efficiency, the contribution of individual tools, and the faithfulness of their reasoning.

Scientific context: Foundation models for CT have multiplied: vision‐language encoders (CT‐CLIP [10], Merlin [8], Jolia [15]), generative models that answer questions about a volume (CT‐CHAT [10], MedGemma 1.5 [11]), large pretrained encoders (Pillar‐0 [12], Curia [13], FlexiCT [14]) and segmentation tools (TotalSegmentator [9]). On their own, they still fail on tasks that require localizing, measuring and then reasoning.

Hence the idea, following ReAct [16], of delegating the reasoning to an LLM that calls these models as tools. Born in chest radiography with MedRAX [7], this paradigm now reaches 3D CT: CT‐Agent [2], 3DMedAgent [5], CT‐Flow [6], INFORM‐CT [3].

The most striking result comes from RadAgent [1]: without any training, an agent orchestrating CT‐CHAT, CT‐CLIP, TotalSeg‐mentator and a general‐purpose 2D VLM (Gemma 3 [19]) outperforms CT‐CHAT on report generation. On DeepTumorVQA [4], a fine‐tuned 2D VLM even beats the best 3D model (66.3% vs. 59.8%), but without measurement tools, all models make up quantitative values.

The problem: each system comes with its own harness (agent loop, tools, prompts), data and metrics, so it is impossible to tell what actually makes the difference. Abdominal CT, less studied than the chest, is an ideal testbed: DeepTumorVQA pro‐vides questions with verifiable answers on 9,262 public volumes, organized by level (recognition, measurement, reasoning), together with a standardized tool environment.

Mission confiée

Research questions:

1. What makes a good agent, and at what cost? Compare harnesses and open‐source LLMs run locally (Qwen3 [17], gptoss

[18], Gemma 4 [22]…) on the same task, and map the accuracy–cost frontier (tokens, latency, GPU memory).

2. Which tools really matter? Systematic ablations and attribution of performance to each tool (segmentation, measurements,

2D or 3D VQA such as MedGemma 1.5 [11], foundation models).

3. Is the agent’s reasoning faithful? When the agent justifies its answer, does it describe what actually determined it?

Possible extension, depending on results and interest: the verifiable answers of DeepTumorVQA make it a natural testbed for reinforcement learning of agents [21] (e.g. GRPO on a small LLM).

Principales activités

Bibliographic study

Programming in PyTorch

Use of distant GPU servers

Scientific presentations and reporting

Ablation studies and benchmarking

Compétences

  • Master 2 in machine learning or final year of engineering school.
  • Good command of Python and PyTorch; taste for rigorous experimentation.
  • A plus: experience with LLMs (Hugging Face, vLLM, tool calling), agent frameworks or evaluation methodology.
  • No prior medical knowledge is required. A first experience with 3D imaging (NIfTI, DICOM, MONAI) is appreciated but can be acquired during the internship.
  • Fluent French or English; comfortable reading scientific papers in English.

Avantages

Subsidized meals

Partial reimbursement of public transport costs

Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)

Possibility of teleworking and flexible organization of working hours

Professional equipment available (videoconferencing, loan of computer equipment, etc.)

Social, cultural and sports events and activities

Access to vocational training

Social security coverage

Rémunération

Gratification

Postuler à cette offre

Partager

Facebook

Linkedin

Twitter

Email

Informations générales

Thème/Domaine : Optimisation, apprentissage et méthodes statistiques

Statistiques (Big data) (BAP E)

Ville : Gif sur Yvette

Centre Inria : Centre Inria de Saclay

Date de prise de fonction souhaitée : 2027-03-01

Durée de contrat : 6 mois

Date limite pour postuler : 2027-02-28

Attention: Les candidatures doivent être déposées en ligne sur le site Inria. Le traitement des candidatures adressées par d'autres canaux n'est pas garanti.

Consignes pour postuler

Sécurité défense :

Ce poste est susceptible d’être affecté dans une zone à régime restrictif (ZRR), telle que définie dans le décret n°2011-1425 relatif à la protection du potentiel scientifique et technique de la nation (PPST). L’autorisation d’accès à une zone est délivrée par le chef d’établissement, après avis ministériel favorable, tel que défini dans l’arrêté du 03 juillet 2012, relatif à la PPST. Un avis ministériel défavorable pour un poste affecté dans une ZRR aurait pour conséquence l’annulation du recrutement.

Politique de recrutement :

Dans le cadre de sa politique diversité, tous les postes Inria sont accessibles aux personnes en situation de handicap.

Contacts

Équipe Inria : OPIS

Recruteur :

Chouzenoux Emilie /

L'essentiel pour réussir

This internship is strictly open to students in the final year of an engineering degree or enrolled in a second-year Master's programme (M2), preferably specializing in machine learning, artificial intelligence, or a closely related field. Applicants should have strong programming skills in Python and PyTorch, a solid background in machine learning, and an interest in rigorous scientific experimentation. Familiarity with large language models (LLMs), agentic AI, or evaluation methodologies would be an advantage. Prior experience in medical imaging is appreciated but not required. Applications from candidates outside the specified academic levels will not be considered.

How to apply: Jobin application + send a CV, M1 and M2 transcripts (or equivalent) and a motivation letter, and . Applications are reviewed on a rolling basis. Incomplete applications are discarded.

Recruitment is subject to the candidate obtaining the required authorization to access the laboratory, which is classified as a Zone à Régime Restrictif (ZRR).

A propos d'Inria

Inria, l'institut national de recherche dans les sciences et technologies du numérique, est en appui de l’État pour les stratégies nationales de recherche et d’innovation du numérique en tant qu'Agence de programmes. Inria mène plus de 300 projets de recherche et d’innovation avec ses 3500 scientifiques, ingénieurs et personnels d’appui, en partenariat avec les universités et l’écosystème numérique (entreprises, entrepreneurs, acteurs publics). Ensemble, nous explorons des domaines clés comme l'intelligence artificielle, la cybersécurité, l’informatique quantique, le Cloud, la transformation numérique de la santé, les jumeaux numériques ou encore les technologies numériques pour la défense. Nous construisons des solutions concrètes telles que des logiciels, des startups technologiques, des partenariats avec les entreprises du tissu national et des formations de pointe. Notre objectif : l’impact scientifique, technologique et industriel au service de la souveraineté numérique de la France.

Plan du site

Contact

Mentions légales

Accessibilité : totalement conforme

Original posting: jobs.inria.fr

Apply on inria.fr (opens in a new tab)Save on FWB