About
Hi👋, I'm Qianli Wang and work as a NLP researcher dedicated to advancing the interpretability and explainability of LLMs at Quality and Usability Lab, Technische Universität Berlin under the supervision of Prof. Dr.-Ing. Sebastian Möller, co-supervised by Dr.-Ing. Nils Feldhus.
News
2026-09-26: Two new papers have been submitted on (1) SAE-based CoT faithfulness metrics; (2) locally sound yet globally insufficient traces in multi-hop reasoning.
2026-09-15: 🤖 Will join Amazon as an Applied Scientist Intern and will work on agentic user simulation.
2026-09-03: Limitations of Automated Simulatability: LLM Simulators Can Bypass Explanations led by Antonin is accepted at BlackboxNLP @ EMNLP 2026!
2026-08-21: 💥 Extremely excited to shared that 6 papers, including 3 (co)-first authored ones, have been accepted at EMNLP 2026 about: (1) multilingual counterfactual explanation alignment (Findings); (2) the impact of quantization on self-explanation quality and faithfulness (Findings); (3) iterative feedback-driven counterfactual generation (Findings);(4) polarity-aware evidence grounding (Main); (5) a meta analysis of NLG evaluation (Main); (6) a tool-augmented drop-in ConvXAI layer (Findings). See you in Budapest 🇭🇺!
2026-07-22: One paper about the limitation of automated simulatability led by Antonin Poché is in submission.
2026-05-22: Three new papers have been submitted on (1) LLM-as-a-Judge circuit discovery; (2) CoT faithfulness meta-evaluation under optimization; (3) multilingual self-explanation alignment.
2026-04-29: Our paper about quantization's effect on factual knowledge recall is finally accepted (Oral) at TrustNLP workshop co-located at ACL 2026!
2026-04-07: 🌟I am thrilled to shared that two papers are accepted at ACL 2026 about (1) multilingual counterfactual evaluation (Main); (2) actionable mechanistic interpretability survey (Findings)! See you in San Diego 🇺🇸
Selected Publications
View All →Parallel Universes, Parallel Languages: A Comprehensive Study on LLM-based Multilingual Counterfactual Example Generation
Qianli Wang, Van Bach Nguyen, Yihong Liu, Fedor Splitt, Nils Feldhus, Christin Seifert, Hinrich Schütze, Sebastian Möller, Vera Schmitt
ACL 2026
Cross-Refine: Improving Natural Language Explanation Generation by Learning in Tandem
Qianli Wang, Tatiana Anikina†, Nils Feldhus†, Simon Ostermann, Sebastian Möller, Vera Schmitt
COLING 2025
Education Background 👨🎓
2024.07-ongoing: Computer Science, Ph.D., Technische Universität Berlin. Supervised by Prof. Dr. Sebastian Möller: Generation, Evaluation, and Application of Natural Language Counterfactual Examples.
2025.09: Visiting PhD Student, University of Copenhagen. Supervised by Prof. Dr. Isabelle Augenstein.
2021.10-2023.10: Computer Science, M.Sc., Technische Universität Berlin. Supervised by Dr.-Ing. Nils Feldhus and Dr.-Ing. Leonhard Hennig: A Singular LLM is all you need for dialogue-based explanation regarding NLP tasks.
2018.10-2021.09: Computer Science, B.Sc., Freie Universität Berlin. Supervised by Prof. Dr. László Kozma: Understanding and implementation of the algorithm for three-coloring in triangle-free planar graphs.
Jobs 🧑💻
2026.09 - 2027.03: Applied Scientist Intern @ Amazon, Berlin, Germany.
2026.06 - 2026.08: Applied Scientist Intern @ Hithink RoyalFlush, Hangzhou, China.
2024.07 - 2026.06: NLP Researcher @ Technische Universität Berlin, Berlin, Germany; funded by VERANDA project.
2024.01 - 2024.05: Machine Learning Researcher @ Aptiv, Wuppertal, Germany.
2023.01 - 2024.01: Research Assistant @ German Research Center for Artificial Intelligence (DFKI) - SLT Group, Berlin, Germany.