Retrieval-Augmented Entity Description Generation for Emerging Entity Linking and Disambiguation
AI & ML interests
ML & NLP
Recent Activity
Cross Context Coreference Resolution
A collection of Italian benchmarks for Large Language Models. See also our Github repo: https://github.com/SapienzaNLP/ita-bench
Efficient and Accurate Coreference Resolution models.
Zero-Shot Example-Based Retrieval Augmentation for Commonsense Question Answering
a Wide-coverage Semantically-annotated Multimodal Dataset
The first family of LLMs pretrained from scratch on Italian.
-
sapienzanlp/Minerva-7B-instruct-v1.0
Text Generation • 7B • Updated • 1.92k • 21 -
sapienzanlp/Minerva-7B-base-v1.0
Text Generation • 7B • Updated • 1.77k • 14 -
sapienzanlp/Minerva-3B-base-v1.0
Text Generation • 3B • Updated • 2.12k • 54 -
sapienzanlp/Minerva-1B-base-v1.0
Text Generation • 1B • Updated • 217 • 28
A blazing fast and lightweight Information Extraction model for Entity Linking and Relation Extraction.
-
ReLiK: Retrieve and LinK, Fast and Accurate Entity Linking and Relation Extraction on an Academic Budget
Paper • 2408.00103 • Published • 23 -
Information Extraction
🏃49Extract entities and relations from text
-
sapienzanlp/relik-entity-linking-large
Updated • 187 • 11 -
sapienzanlp/relik-entity-linking-base
Updated • 608 • 4
An unofficial collection of Italian LLMs developed by iGenius.
Machine Translation (MT) metrics designed explicitly to scrutinize the MT meta-evaluation process’s accuracy, robustness, and fairness.
Retrieval-Augmented Entity Description Generation for Emerging Entity Linking and Disambiguation
a Wide-coverage Semantically-annotated Multimodal Dataset
Cross Context Coreference Resolution
The first family of LLMs pretrained from scratch on Italian.
-
sapienzanlp/Minerva-7B-instruct-v1.0
Text Generation • 7B • Updated • 1.92k • 21 -
sapienzanlp/Minerva-7B-base-v1.0
Text Generation • 7B • Updated • 1.77k • 14 -
sapienzanlp/Minerva-3B-base-v1.0
Text Generation • 3B • Updated • 2.12k • 54 -
sapienzanlp/Minerva-1B-base-v1.0
Text Generation • 1B • Updated • 217 • 28
A collection of Italian benchmarks for Large Language Models. See also our Github repo: https://github.com/SapienzaNLP/ita-bench
A blazing fast and lightweight Information Extraction model for Entity Linking and Relation Extraction.
-
ReLiK: Retrieve and LinK, Fast and Accurate Entity Linking and Relation Extraction on an Academic Budget
Paper • 2408.00103 • Published • 23 -
Information Extraction
🏃49Extract entities and relations from text
-
sapienzanlp/relik-entity-linking-large
Updated • 187 • 11 -
sapienzanlp/relik-entity-linking-base
Updated • 608 • 4
Efficient and Accurate Coreference Resolution models.
An unofficial collection of Italian LLMs developed by iGenius.
Zero-Shot Example-Based Retrieval Augmentation for Commonsense Question Answering
Machine Translation (MT) metrics designed explicitly to scrutinize the MT meta-evaluation process’s accuracy, robustness, and fairness.