Deutsch Intern
    Natural Language Processing

    Open position: Postdoc to work on aligning LLMs

    09/01/2023

    We are looking for a postdoc to join our group! The position is bound to the project EQUIFAIR (Equitably Fair and Trustworthy Language Technology), funded by the Alcatel-Lucent Stiftung and focuses on alignment of large language models (LLMs): hallucinations and societal biases, to be addressed in a (massively) multilingual context and in an explainable/interpretable manner.

    We are hiring a postdoctoral NLP researcher to work on the project EQUIFAIR (Equitably Fair and Trustworthy Language Technology), funded by the Alcatel-Lucent Stiftung as part of the “Responsible AI” program. The project targets two major issues related to alignment of large language models (LLMs): hallucinations and societal biases, to be addressed in a (massively) multilingual context (i.e., across a wide variety of natural languages) and in an explainable/interpretable manner.

    The applicants should have a doctoral degree in Computer Science, Mathematics, or a related discipline. Substantial experience in modern NLP (i.e., LLM-based) research is expected, as well as prior publications at top-tier NLP/ML venues (e.g., ACL, EMNLP, NAACL, NeurIPS, ICLR, ICML), demonstrating ability to conduct research at the highest level. Prior experience in any of the subareas relevant to the project, namely fair and trustworthy NLP (hallucinations, biases), multilinguality, and/or explainable/interpretable AI, is a plus.

    The position is available for two years; the project will start as soon as the suitable candidate for this position is found. The remuneration will be at the level E14 of the German federal wage agreement scheme (TV-L).

    Applications: per email to goran.glavas@uni-wuerzburg.de latest by 30.09.2023. 

    The full job annoncement is available here.