-
BERT-as-a-Judge: A Robust Alternative to Lexical Methods for Efficient Reference-Based LLM Evaluation
Paper • 2604.09497 • Published • 26 -
artefactory/BERTJudge
Text Classification • 0.2B • Updated • 60 • 3 -
artefactory/BERTJudge-Free-CR
Text Classification • 0.2B • Updated • 4 • 1 -
artefactory/BERTJudge-Formatted-QCR
Text Classification • 0.2B • Updated • 61 • 1
AI & ML interests
NLP, Information Retrieval, Computer Vision, Uncertainty Estimation, Trustworthy AI, Bias Estimation, Unbalanced ML, Choice Modeling, Time Series
Recent Activity
View all activity
Papers
BERT-as-a-Judge: A Robust Alternative to Lexical Methods for Efficient Reference-Based LLM Evaluation
Learned Hallucination Detection in Black-Box LLMs using Token-level Entropy Production Rate
Related paper: "Should We Still Pretrain Encoders with Masked Language Modeling?"
Related paper: "Towards Trustworthy Reranking: A Simple yet Effective Abstention Mechanism" (accepted at TMLR 2024)
Suite of Encoder models EuroBERT
Related paper: "Is Preference Alignment Always the Best Option to Enhance LLM-Based Translation? An Empirical Analysis" (accepted at WMT 2024)
-
Is Preference Alignment Always the Best Option to Enhance LLM-Based Translation? An Empirical Analysis
Paper • 2409.20059 • Published • 16 -
artefactory/ALMA-13B-LoRA
Text Generation • 13B • Updated • 5 -
artefactory/ALMA-13B-LoRA-SFT-xCOMET-QE-Multi
Text Generation • 13B • Updated • 3 -
artefactory/ALMA-13B-LoRA-SFT-xCOMET-QE-Multi-No-Base
Text Generation • 13B • Updated • 1
-
BERT-as-a-Judge: A Robust Alternative to Lexical Methods for Efficient Reference-Based LLM Evaluation
Paper • 2604.09497 • Published • 26 -
artefactory/BERTJudge
Text Classification • 0.2B • Updated • 60 • 3 -
artefactory/BERTJudge-Free-CR
Text Classification • 0.2B • Updated • 4 • 1 -
artefactory/BERTJudge-Formatted-QCR
Text Classification • 0.2B • Updated • 61 • 1
Related paper: "Should We Still Pretrain Encoders with Masked Language Modeling?"
Suite of Encoder models EuroBERT
Related paper: "Towards Trustworthy Reranking: A Simple yet Effective Abstention Mechanism" (accepted at TMLR 2024)
Related paper: "Is Preference Alignment Always the Best Option to Enhance LLM-Based Translation? An Empirical Analysis" (accepted at WMT 2024)
-
Is Preference Alignment Always the Best Option to Enhance LLM-Based Translation? An Empirical Analysis
Paper • 2409.20059 • Published • 16 -
artefactory/ALMA-13B-LoRA
Text Generation • 13B • Updated • 5 -
artefactory/ALMA-13B-LoRA-SFT-xCOMET-QE-Multi
Text Generation • 13B • Updated • 3 -
artefactory/ALMA-13B-LoRA-SFT-xCOMET-QE-Multi-No-Base
Text Generation • 13B • Updated • 1