5 papers
Two-dimensional early exit optimisation of LLM inference
Jan Hůla, David Adamczyk, Tomáš Filip +2
We introduce a two-dimensional (2D) early exit strategy that coordinates layer-wise and sentence-wise exiting for classification tasks in large language models. By processing input…
Geometric Reasoning in the Embedding Space
Jan Hůla, David MojžÃÅ¡ek, JiÅà JaneÄek +2
In this contribution, we demonstrate that Graph Neural Networks and Transformers can learn to reason about geometric constraints. We train them to predict spatial position of point…
Minimizing the Weighted Number of Tardy Jobs: Data-Driven Heuristic for Single-Machine Scheduling
Nikolai Antonov, PrÄmysl Šůcha, Mikoláš Janota +1
Existing research on single-machine scheduling is largely focused on exact algorithms, which perform well on typical instances but can significantly deteriorate on certain regions…
BenCzechMark : A Czech-centric Multitask and Multimetric Benchmark for Large Language Models with Duel Scoring Mechanism
Martin Fajcik, Martin Docekal, Jan Dolezal +15
We present BenCzechMark (BCM), the first comprehensive Czech language benchmark designed for large language models, offering diverse tasks, multiple task formats, and multiple eval…
Neural Approaches to SAT Solving: Design Choices and Interpretability
David MojžÃÅ¡ek, Jan Hůla, Ziwei Li +2
In this contribution, we provide a comprehensive evaluation of graph neural networks applied to Boolean satisfiability problems, accompanied by an intuitive explanation of the mech…