LEMUR and Mean Centering for Late-Interaction Retrieval in txtai
• 9
AI & ML interests
Applying machine learning to solve everyday problems
Recent Activity
View all activity
Articles
Add Sentence Transformers usage
🚀 1
1
#1 opened 6 days ago
by 
tomaarsen
Add Sentence Transformers usage
🚀 1
3
#1 opened 10 days ago
by tomaarsen
davidmezzetti
posted an update 13 days ago
Post
1722
Exciting addition coming with the next txtai release: LEMUR for ColBERT-style Late-Interaction Retrieval! 🎉
Contributor @Morgan-coded introduced LEMUR to txtai, making it, as far as we know, the first framework to incorporate LEMUR for late-interaction retrieval using standard, fixed-vector indexes.
Key benefits:
🚀 Significant boost: 49–62% higher NDCG@10 than 2,048-dimensional MUVERA
💾 5x less storage: 2,048 dimensions vs. MUVERA’s default 10,240
📐 Better geometry: Optional batch mean centering addresses anisotropy in token embeddings
A promising step toward making ColBERT-style retrieval more practical with conventional vector search.
Read the full breakdown: https://huggingface.co/blog/NeuML/txtai-lemur
Contributor @Morgan-coded introduced LEMUR to txtai, making it, as far as we know, the first framework to incorporate LEMUR for late-interaction retrieval using standard, fixed-vector indexes.
Key benefits:
🚀 Significant boost: 49–62% higher NDCG@10 than 2,048-dimensional MUVERA
💾 5x less storage: 2,048 dimensions vs. MUVERA’s default 10,240
📐 Better geometry: Optional batch mean centering addresses anisotropy in token embeddings
A promising step toward making ColBERT-style retrieval more practical with conventional vector search.
Read the full breakdown: https://huggingface.co/blog/NeuML/txtai-lemur
- 32 replies
davidmezzetti
published an article 14 days ago
Article
LEMUR and Mean Centering for Late-Interaction Retrieval in txtai
NeuML
• • 9 davidmezzetti
updated 3
collections 26 days ago
davidmezzetti
updated a
model 26 days ago
davidmezzetti
published a
model 26 days ago
davidmezzetti
updated a
dataset 29 days ago
davidmezzetti
updated a
collection about 1 month ago
davidmezzetti
posted an update about 1 month ago
Post
654
The year is 1899. It's still a horse and buggy world. Einstein hasn't published his famous annus mirabilis papers setting the foundation for Physics as we understand it today. The world is advancing at a rapid pace roaring into the 1900s. What if AI models were trained in 1899 and given to the best minds of the day? Could there have been alternate paths for discovery on par or perhaps even ahead of where we are in 2026?
Introducing the new H.G. BERT Small series of models. This is a 22.7M parameter BERT encoder-only model trained from scratch ONLY on Historical English Books from 1700 - 1899.
https://huggingface.co/blog/NeuML/hgbert-small
Introducing the new H.G. BERT Small series of models. This is a 22.7M parameter BERT encoder-only model trained from scratch ONLY on Historical English Books from 1700 - 1899.
https://huggingface.co/blog/NeuML/hgbert-small
davidmezzetti
published an article about 1 month ago
Article
H.G. BERT Small: AI like it's 1899
NeuML
• • 2 davidmezzetti
updated a
model about 1 month ago