Challenge to distinguish whether a sentence from a news article expresses the subjective view of the author behind it or presents an objective view on the covered topic

Post-competition solution to the first task of the TalentCLEF 2025 challenge (NLP Master's final project).

CLEF2025_Task4a is the official repository for Subtask 4a of the CLEF2025 CheckThat Lab. It provides a complete pipeline for building and evaluating multi-label classifiers to detect scientific discourse in Twitter data. This project supports reproducible research in social media analysis and NLP.

Our submission on Task 2 on Multilingual Claim Normalization, achieving achieved top-three positions in 15 of the 20 languages.

Research project for CLEF 2025 Touché Lab on “Advertisements in Retrieval-Augmented Generation.” Covers Subtask 1 (ad generation via RAG pipeline with FAISS, CrossEncoder, Qwen) and Subtask 2 (ad detection using fine-tuned DeBERTa and RoBERTa models). Ranked 2nd and 4th internationally, presented at CLEF 2025 in Madrid.