Scholar
Carel van Niekerk
Google Scholar ID: Asit7LQAAAAJ
Heinrich-Heine University, Düsseldorf
Citations & Impact
Citations
671
H-index
11
i10-index
14
Publications
20
Co-authors
16
list available
Publications
Resume (English only)
Academic Achievements
- Co-authored the 2020 SIGDial best paper 'TripPy'.
- Paper 'Less is More: Local Intrinsic Dimensions of Contextual Language Models' accepted at NeurIPS 2025.
- Released preprint 'Post-Training Large Language Models via Reinforcement Learning from Self-Feedback' (July 2025), introducing RLSF—a novel self-feedback reinforcement learning approach.
- Paper 'A Confidence-based Acquisition Model for Self-supervised Active Learning and Label Correction' published in TACL and to be presented at ACL 2025.
- Published at top venues including EMNLP, ACL, and IEEE, with a focus on generative modeling and trustworthy AI.
Background
- Currently a Postdoctoral Researcher in the Dialogue Systems and Machine Learning group at Heinrich Heine University, Düsseldorf, under Prof. Milica Gašić.
- Research interests focus on understanding uncertainty in Large Language Models (LLMs), generative AI, trustworthy AI, and reinforcement learning for LLM fine-tuning.
- Has 10 years of combined academic and industry experience in machine learning, deep learning, and natural language processing.
- Previously worked as an AI consultant, applying state-of-the-art AI methods to real-world problems.
- Aims to make dialogue systems smarter and more reliable through deeper insights into LLMs.
