Josef Kuchar

According to our database, Josef Kuchar authored at least 7 papers between 2025 and 2026.

Collaborative distances:

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

2026

Language Models Learn Universal Representations of Numbers and Here's Why You Should Care.

Michal Stefánik, Timothee Mickus, Marek Kadlcík, Bertram Højer, Michal Spiegel, Raúl Vázquez, Aman Sinha, Josef Kuchar, Philipp Mondorf, Pontus Stenetorp

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2026

2025

Unravelling the Mechanisms of Manipulating Numbers in Language Models.

Michal Stefánik, Timothee Mickus, Marek Kadlcík, Bertram Højer, Michal Spiegel, Raúl Vázquez, Aman Sinha, Josef Kuchar, Philipp Mondorf

CoRR, October, 2025

Can Out-of-Distribution Evaluations Uncover Reliance on Shortcuts? A Case Study in Question Answering.

Michal Stefánik, Timothee Mickus, Marek Kadlcík, Michal Spiegel, Josef Kuchar

CoRR, August, 2025

VectorEdits: A Dataset and Benchmark for Instruction-Based Editing of Vector Graphics.

Josef Kuchar, Marek Kadlcík, Michal Spiegel, Michal Stefánik

CoRR, June, 2025

Attend or Perish: Benchmarking Attention in Algorithmic Reasoning.

Michal Spiegel, Michal Stefánik, Marek Kadlcík, Josef Kuchar

CoRR, March, 2025

Can Out-of-Distribution Evaluations Uncover Reliance on Prediction Shortcuts? A Case Study in Question Answering.

Michal Stefánik, Timothee Mickus, Michal Spiegel, Marek Kadlcík, Josef Kuchar

Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2025, 2025

Pre-trained Language Models Learn Remarkably Accurate Representations of Numbers.

Marek Kadlcík, Michal Stefánik, Timothee Mickus, Josef Kuchar, Michal Spiegel

Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, 2025