Internship Opportunity Self-Improving Large Language Models through Multi-Agent Reasoning

# | | | | --- | --- | | | | | --- | | Internship Opportunity Self-Improving Large Language Models through Multi-Agent Reasoning | |Back| | || --- | --- || | | | --- | | The Bruno Kessler Foundation (FBK) is a leading research institution based in Trento, Italy. With over 450 researchers distributed across 12 Research Centers, FBK promotes excellence in science and technology across both technical and social domains. The internship is offered within the Intelligent Digital Agents (IDA) Research Unit of the Digital Health & Wellbeing Center. The unit focuses on developing flexible, ethically grounded technologies and digital assistants, with particular attention to multi-agent systems, reasoning, human-centered AI, and digital health applications. More information can be found at **Planned activities** The Research Unit is looking for a Master’s student interested in pursuing an internship experience on **Self-Improving Large Language Models through Multi-Agent Reasoning**. The internship will explore how a general-purpose Large Language Model, such as Gemma or similar open-weight models, can improve its reasoning performance without relying on external training data. The work will build on the Pool of Experts (PoE) framework, a multi-agent architecture in which several expert agents analyze a problem from different perspectives and cooperate to produce a final decision. The main goal of the internship is to investigate whether PoE can be used not only as an inference-time reasoning framework, but also as a mechanism for model self-improvement. Starting from a generalist LLM, the intern will study how the multi-agent system can be used to improve model performance without introducing new annotated datasets. Main activities include: * Exploring self-improvement strategies based on internal model outputs, such as self-consistency, self-reflection, critique, debate, and self-distillation; * Designing experiments in which PoE agents generate, evaluate, revise, or select improved reasoning paths; * Investigating whether synthetic traces produced by the model itself can support lightweight adaptation or improved inference strategies; * Comparing baseline single-model performance with PoE-based self-enhancement strategies. This internship offers experience at the frontier of Large Language Models, multi-agent systems, self-improving AI, and reasoning-oriented model evaluation. **Requirements** * Enrollment in a Master’s degree in one of the following fields: Computer Science, Data Science, Artificial Intelligence, Cognitive Sciences, or related disciplines; * Fluent knowledge of English, minimum level B2; * Strong interpersonal and teamwork skills; * Good adaptability, initiative, and autonomy; * Proficiency in Python; * Familiarity with Large Language Models, generative AI, and model training; * Interest in multi-agent systems, reasoning, model evaluation, or efficient adaptation techniques. Additional knowledge of open-weight LLMs, LoRA, prompt engineering, or evaluation benchmarks will be considered a plus. **Internship information** * Internship start date: second semester 2026; * Internship experience duration: 6 months; * Opportunity reserved for Master’s students; * The internship experience will take place at FBK; * We offer support in finding accommodation at affiliated facilities, in the case of off-site candidates; * We offer the possibility to use the internal canteen service; * Possible recognition of participation allowance. **How to Apply** Interested candidates are asked to send their application by email to: **Patrizio Bellan** – Visualizzare la mail su click.appcast.io The application must include the following documents in .pdf format: * curriculum vitae; * motivation letter. **The motivation letter is required to apply.** For any further information, please contact the Human Resources Department at: Visualizzare la mail su click.appcast.io Application deadline: the notice will be withdrawn when the desired applications are reached. | | #J-18808-Ljbffr