Mikolaj Langner

According to our database, Mikolaj Langner authored at least 4 papers between 2025 and 2026.

Collaborative distances:

• Dijkstra number of five.
• Erdős number of four.

2026

What properties of reasoning supervision are associated with improved downstream model quality?

Mikolaj Langner, Dzmitry Pihulski, Jan Eliasz, Michal Rajkowski, Przemyslaw Kazienko, Maciej Piasecki, Jan Kocon, Teddy Ferdinan

CoRR, May, 2026

Breaking the Illusion of Reasoning in Polish LLMs: Quality over Quantity of Thought.

Dzmitry Pihulski, Mikolaj Langner, Jan Eliasz, Przemyslaw Kazienko, Jan Kocon, Teddy Ferdinan

Proceedings of the Findings of the Association for Computational Linguistics: EACL 2026, 2026

2025

Divide, Cache, Conquer: Dichotomic Prompting for Efficient Multi-Label LLM-Based Classification.

Mikolaj Langner, Jan Eliasz, Ewa Rudnicka, Jan Kocon

Proceedings of the IEEE International Conference on Data Mining, 2025

AggTruth: Contextual Hallucination Detection Using Aggregated Attention Scores in LLMs.

Piotr Matys, Jan Eliasz, Konrad Kielczynski, Mikolaj Langner, Teddy Ferdinan, Jan Kocon, Przemyslaw Kazienko

Proceedings of the Computational Science - ICCS 2025 Workshops, 2025