teanalab-NACER/NACER: Benchmark and Neural Architecture for Conversational Entity Retrieval from a Knowledge Graph

Source code of Neural Architecture for Conversational Entity Retrieval (NACER), a feature-based neural architecture for KG candidate entity ranking in Conversational Entity Retrieval from DBpedia and the baseline models (BM25F, KV-MemNN, GENRE and LLaMa) used in Mona Zamiri, Yao Qiang, Fedor Nikolaev, Dongxiao Zhu, Alexander Kotov. Benchmark and Neural Architecture for Conversational Entity Retrieval from a Knowledge Graph. In Proceedings of the 2024 ACM Web Conference 2024 (WWW ’24).