Data Science Group at UPB
Change the repository type filter
All
Repositories list
365 repositories
- A hybrid LLM-based framework that performs parallel mention expansion using multilingual and monolingual LLMs, then applies a similarity-based mechanism to select the most contextually appropriate expansion. The selected expansion is used to improve entity disambiguation model in low-resource languages, such as Indonesian.
owlapy
Public- OWL Class Expressions Learning in Python
tentris-research-project
PublicTentris is a tensor-based triplestore that natively supports worst-case optimal joins. This is the research version of Tentris. A commercial version is available at https://github.com/tentris/tentris.hypertrie
PublicA monolithic index that supports worst-case optimal joins (WCOJ) by providing all collation orders in a single redundancy eliminating datastructure.gerbil
PublicGERBIL - General Entity annotatoR BenchmarkIGUANA
PublicIGUANA is a benchmark execution framework for RDF triple stores and quad storesmetall-ffi
PublicKGSUMM-PG
Publicdice-template-library
Publicmetall
PublicANTS
PublicThis repository contains the implementation of ANTS, a research paper accepted at ESWC 2025 (Research Track). ANTS addresses the challenges of abstractive entity summarization in Knowledge Graphs (KGs) by generating optimal summaries that integrate existing triples with inferred (absent) triples.Benchmarking-KE
PublicBenchmarking-LLM
PublicWHALE-embeddings
Publicserd
Publicembeddings.cc
Publicbenckmarking-KE-w3id.org
PublicOpus-MT
PublicNEBULA
Public