Master Thesis in Multi-Modal Retrieval-Augmented Generation

Bosch Group; Renningen; Vor Ort; EUR 40.000 - 60.000; Vollzeit

Zusammenfassung

A leading technology firm in Germany is offering a master thesis opportunity to develop a multi-modal retrieval-augmented generation system for multi-hop question answering. This project involves integrating various data types and benchmarking models. Ideal candidates should be pursuing a Master’s in Computer Science or Mathematics, have experience with machine learning, and be proficient in English. This 6-month thesis is a great chance to contribute to cutting-edge research.

Qualifikationen

• Motivated team player with strong communication skills.
• Passionate about research and independent problem solving.
• Very good in English.

Aufgaben

• Develop a multi-modal RAG model integrating combined multi-modal information.
• Benchmark existing baselines.

Kenntnisse

large language models (LLMs)

vision-language models (VLMs)

deep learning frameworks (PyTorch)

multi-modal learning

retrieval models

Ausbildung

Master studies in Computer Science or Mathematics

Jobbeschreibung

Company Description

At Bosch, we shape the future by inventing high-quality technologies and services that spark enthusiasm and enrich people’s lives. Our promise to our associates is rock-solid: we grow together, we enjoy our work, and we inspire each other. Join in and feel the difference.

The Robert Bosch GmbHis looking forward to your application!

Job Description

This master thesis aims to develop a multi-modal retrieval-augmented generation (RAG) system capable of integrating textual, visual and structured knowledge to perform robust multi-hop question answering. Such a system has practical applications in areas like manufacturing, autonomous driving and multimedia search.

• In your master thesis you will develop a multi-modal RAG model that integrates combined multi-modal information. You will also benchmark existing baselines.

For the master thesis topic, one of the following research questions can be chosen:

• How can multi-modal data (text, images, knowledge graphs) be effectively integrated within a retrieval-augmented generation framework for multi-hop question answering?
• How robust is the multi-modal RAG model under realistic domain shifts or noisy/missing modalities?

Qualifications

• Education: Master studies in the field of Computer Science, Mathematics or comparable
• Experience and Knowledge : in large language models (LLMs), vision-language models (VLMs) and knowledge graphs; experience or willingness to learn deep learning frameworks (PyTorch); prior knowledge of multi-modal learning and retrieval models is preferred
• Personality and Working Practice: you are a motivated team player with strong communication skills
• Enthusiasm: passion for research and independent problem solving
• Languages: very good in English

Additional Information

Start:according to prior agreement
Duration:6 months

Requirement for this thesis is the enrollment at university. Please attach your CV, transcript of records, examination regulations and if indicated a valid work and residence permit.

Diversity and inclusion are not just trends for us but are firmly anchored in our corporate culture. Therefore, we welcome all applications, regardless of gender, age, disability, religion, ethnic origin or sexual identity.

Need further information about the job?
Hongkuan Zhou (Functional Department)
[phone]
Lavdim Halilaj (Functional Department)
[phone]

#LI-DNI