Postdoc: Llm-powered Conversational Agents

Wouldn't it be great if conversational agents were truly trustworthy? LLM-powered conversational agents excel at various roles and tasks.. However, they can be brittle and lack robustness when used in real-world scenarios, where requests can be inappropriate, beyond the assistant's mission, or simply ambiguous.

If you agree it’s a key prerequisite for their acceptability and feel ready to tackle these challenges, you might want to join us for a 12 months CDD around building agents that are trustworthy-by-design

Within the framework of the UTTER EU project, you would work on one or several of these topics: i] managing uncertainty and ambiguity ii] improving the use of conversational context iii] ensuring the safety and alignment of LLMs.

You should be proficient with Large Language Models, including their training and fine-tuning procedures. Familiarity with PyTorch and alignment techniques (e.g. RLHF) is also necessary. You should be comfortable conducting experiments autonomously in an HPC environment.

**Contacts**: Laurent Besacier & Jos Rozen

Application instructions

About NAVER LABS

NAVER is the #1 Internet portal in Korea with activities that span a wide range of businesses including search, commerce, content, financial and cloud platforms.

NAVER LABS, co-located in Korea and France, is the organization dedicated to preparing NAVER’s future. NAVER LABS Europe is located in a spectacular setting in Grenoble, in the heart of the French Alps. Scientists at NAVER LABS Europe are empowered to pursue long-term research problems that, if successful, can have significant impact and transform NAVER. We take our ideas as far as research can to create the best technology of its kind. Active participation in the academic community and collaborations with world-class public research groups are, among others, important tools to achieve these goals. Teamwork, focus and persistence are important values for us.

NAVER LABS Europe is an equal opportunity employer.