-
artefactory/ledger-long-context-KPI-QA
Viewer • Updated • 115k • 3.3k • 12 -
artefactory/ledger-long-context-multi-kpi
Viewer • Updated • 5k • 3.07k • 12 -
artefactory/ledger-market-sentiment
Viewer • Updated • 449k • 272 • 10 -
LEDGER: A Long-Context Benchmark of Corporate Annual Reports for Grounded Financial Retrieval and Extraction
Paper • 2606.13100 • Published • 1
ArGiMi
AI & ML interests
None defined yet.
Recent Activity
ArGiMi is a collaborative project with 5 members: Mistral.ai, Giskard.ai, Artefact, INA and BnF, with general objective to enrich the French artificial intelligence ecosystem and is supported by the France2030 program.
Together, we are building a collection of open source resources. Each partner contributes to the project: Mistral is in charge of creating “foundational” and “fine-tuned” AI models. Artefact connects these models to the real world with industrial corpora and application tools such as RAG. Giskard is responsible for quality assurance: rigorous evaluation of models, testing, and risk mitigation strategies. Finally, INA and BnF are bringing our cultural heritage into the era of generative AI by preparing their precious collections for these new technologies.
Most of these resources will be available on this page.
-
BERT-as-a-Judge: A Robust Alternative to Lexical Methods for Efficient Reference-Based LLM Evaluation
Paper • 2604.09497 • Published • 29 -
artefactory/BERTJudge
Text Classification • 0.2B • Updated • 129 • 9 -
artefactory/BERTJudge-Free-CR
Text Classification • 0.2B • Updated • 14 • 1 -
artefactory/BERTJudge-Formatted-CR
Text Classification • 0.2B • Updated • 16
-
artefactory/ledger-long-context-KPI-QA
Viewer • Updated • 115k • 3.3k • 12 -
artefactory/ledger-long-context-multi-kpi
Viewer • Updated • 5k • 3.07k • 12 -
artefactory/ledger-market-sentiment
Viewer • Updated • 449k • 272 • 10 -
LEDGER: A Long-Context Benchmark of Corporate Annual Reports for Grounded Financial Retrieval and Extraction
Paper • 2606.13100 • Published • 1
-
BERT-as-a-Judge: A Robust Alternative to Lexical Methods for Efficient Reference-Based LLM Evaluation
Paper • 2604.09497 • Published • 29 -
artefactory/BERTJudge
Text Classification • 0.2B • Updated • 129 • 9 -
artefactory/BERTJudge-Free-CR
Text Classification • 0.2B • Updated • 14 • 1 -
artefactory/BERTJudge-Formatted-CR
Text Classification • 0.2B • Updated • 16