Causal information extraction with LLMs

Organisatsiooni nimi
Tartu NLP
Kokkuvõte
With the recent increase in performance and widespread use of LLMs, many avenues are possible for their implementation as support tools. Since processing large texts and reading between the lines (literally) can be time-consuming for humans, couldn't this process be assisted? The goal of this thesis is to investigate the efficacy of LLM architectures in extracting "Why?/How?"-questions based on information from textual data, with different possible procedures.
Lõputöö kaitsmise aasta
2024-2025
Juhendaja
Giacomo Magnifico
Suhtlemiskeel(ed)
inglise keel
Nõuded kandideerijale
Tase
Bakalaureus, Magister
Märksõnad
#ai, #causal_relation, #llm, #text_analysis, #simplification

Kandideerimise kontakt

 
Nimi
Giacomo Magnifico
Tel
E-mail
giacomo.magnifico@ut.ee