Hallucination free RAG and out SOTA state-of-the-art extractors
AI & ML interests
Explainable AI, Rule-based models, Rule learning with LLMs, Hallucination detection, Fact checking LLMs
Recent Activity
This Collection contains our small, Ettin-encoder (https://arxiv.org/abs/2507.11412) based models trained on synthetic and RagTruth data.
-
KRLabsOrg/tinylettuce-ettin-17m-en-bioasq
Token Classification • 16.9M • Updated • 4 • 8 -
KRLabsOrg/tinylettuce-ettin-68m-en-bioasq
Token Classification • 68.4M • Updated • 65 • 2 -
KRLabsOrg/tinylettuce-ettin-32m-en-bioasq
Token Classification • 32M • Updated • 4 • 1 -
KRLabsOrg/tinylettuce-ettin-68m-en
Token Classification • 68.4M • Updated • 165 • 2
This collection includes our hungarian models using the recently released multilingual ModernBERT models
This collection includes our translated training data that we've used to create multilingual hallucination detection models.
LettuceDetect v1. Trained ModernBERT (base and large) for detection hallucinations in LLM responses. The models are trained as token classifications.
-
KRLabsOrg/lettucedect-base-modernbert-en-v1
Token Classification • 0.1B • Updated • 9.93k • • 19 -
KRLabsOrg/lettucedect-large-modernbert-en-v1
Token Classification • 0.4B • Updated • 3.7k • • 30 -
LettuceDetect: A Hallucination Detection Framework for RAG Applications
Paper • 2502.17125 • Published • 14 -
LettuceDetect
🥬7Let Us Detect your hallucinations! Demo for our framework.
Prune tool output with our fine-tuned Qwen 2B model
These are our EuroBERT fine-tunes on our translated RAGTruth datasets.
-
KRLabsOrg/lettucedect-610m-eurobert-de-v1
Token Classification • 0.6B • Updated • 46 • 1 -
KRLabsOrg/lettucedect-210m-eurobert-de-v1
Token Classification • 0.2B • Updated • 103 -
KRLabsOrg/lettucedect-610m-eurobert-fr-v1
Token Classification • 0.6B • Updated • 11 • 1 -
KRLabsOrg/lettucedect-610m-eurobert-cn-v1
Token Classification • 0.6B • Updated • 25 • 1
Hallucination free RAG and out SOTA state-of-the-art extractors
LettuceDetect v1. Trained ModernBERT (base and large) for detection hallucinations in LLM responses. The models are trained as token classifications.
-
KRLabsOrg/lettucedect-base-modernbert-en-v1
Token Classification • 0.1B • Updated • 9.93k • • 19 -
KRLabsOrg/lettucedect-large-modernbert-en-v1
Token Classification • 0.4B • Updated • 3.7k • • 30 -
LettuceDetect: A Hallucination Detection Framework for RAG Applications
Paper • 2502.17125 • Published • 14 -
LettuceDetect
🥬7Let Us Detect your hallucinations! Demo for our framework.
This Collection contains our small, Ettin-encoder (https://arxiv.org/abs/2507.11412) based models trained on synthetic and RagTruth data.
-
KRLabsOrg/tinylettuce-ettin-17m-en-bioasq
Token Classification • 16.9M • Updated • 4 • 8 -
KRLabsOrg/tinylettuce-ettin-68m-en-bioasq
Token Classification • 68.4M • Updated • 65 • 2 -
KRLabsOrg/tinylettuce-ettin-32m-en-bioasq
Token Classification • 32M • Updated • 4 • 1 -
KRLabsOrg/tinylettuce-ettin-68m-en
Token Classification • 68.4M • Updated • 165 • 2
Prune tool output with our fine-tuned Qwen 2B model
This collection includes our hungarian models using the recently released multilingual ModernBERT models
These are our EuroBERT fine-tunes on our translated RAGTruth datasets.
-
KRLabsOrg/lettucedect-610m-eurobert-de-v1
Token Classification • 0.6B • Updated • 46 • 1 -
KRLabsOrg/lettucedect-210m-eurobert-de-v1
Token Classification • 0.2B • Updated • 103 -
KRLabsOrg/lettucedect-610m-eurobert-fr-v1
Token Classification • 0.6B • Updated • 11 • 1 -
KRLabsOrg/lettucedect-610m-eurobert-cn-v1
Token Classification • 0.6B • Updated • 25 • 1
This collection includes our translated training data that we've used to create multilingual hallucination detection models.