https://scidar.kg.ac.rs/handle/123456789/8214 Fri, 06 Nov 2026 19:46:40 GMT 2026-11-06T19:46:40Z https://scidar.kg.ac.rs/handle/123456789/23431 Title: Hitting-Time Index of Trees: Bounds, Relationships, and Exact Formulas Authors: Liang, Jinxia; He, Mengya; Xing, Bingde; Mao, Yaping; Furtula, Boris Abstract: The hitting-time index is a recently introduced topological molecular descriptor based on graph distance. Its lower and upper bounds for trees, in terms of other graph invariants, are derived. In particular, the hitting-time index is related to the eccentric connectivity index, total eccentricity, eccentric distance sum, and degree distance index. Additionally, a closed formula for the hitting-time index of double brooms is obtained. Thu, 01 Jan 2026 00:00:00 GMT https://scidar.kg.ac.rs/handle/123456789/23431 2026-01-01T00:00:00Z https://scidar.kg.ac.rs/handle/123456789/23430 Title: Revisiting the Diminished Sombor Index Authors: Akbari, Saieed; Ali, Akbar; Furtula, Boris; Movahedi, Fateme; Moghadam, Maryam Rahmani Thu, 01 Jan 2026 00:00:00 GMT https://scidar.kg.ac.rs/handle/123456789/23430 2026-01-01T00:00:00Z https://scidar.kg.ac.rs/handle/123456789/23429 Title: Diminished Sombor Index Authors: Movahedi, Fateme; Gutman, Ivan; Redžepović, Izudin; Furtula, Boris Thu, 01 Jan 2026 00:00:00 GMT https://scidar.kg.ac.rs/handle/123456789/23429 2026-01-01T00:00:00Z https://scidar.kg.ac.rs/handle/123456789/23366 Title: Comparing QLoRA fine-tuning of a general-purpose and a mathema-tically specialized LLM on Serbian mathematics competition tasks Authors: Svičević, Marina; Pavković, Miloš; Vučićević, Nemanja; Milenković, Aleksandar; Milutinović, Aleksandar Abstract: This paper examines the effect of QLoRA fine-tuning of locally executable large language models on mathematics competition tasks for high school students in Serbia. The main focus is on comparing the general purpose model Mistral-7BInstruct and the mathematically specialized model Mathstral-7B, in order to determine whether prior specialization for mathematical reasoning affects the success of fine-tuning on a small, domain-specific corpus of tasks in Serbian. For the purposes of the study, an automatic evaluation framework was applied, in which generated solutions were assessed according to several criteria, including final answer accuracy, logical coherence of the solution, and explanation quality. The results indicate different effects of fine-tuning for the observed models: the general-purpose model showed a slight performance decline, while the mathematically specialized model showed an improvement. This outcome suggests that models previously adapted to mathematical reasoning may benefit more from domain-specific fine-tuning, while also confirming that Serbian mathematics competition tasks remain a highly demanding test for locally executable large language models. Thu, 01 Jan 2026 00:00:00 GMT https://scidar.kg.ac.rs/handle/123456789/23366 2026-01-01T00:00:00Z