Online soft compression models for RAG. We release the models with llama-1B as compressor.
AI & ML interests
3D vision, speech, NLP, neural retrieval, computer vision, LLM, FMs, visual localization, robotics
PISCO models are compression models for RAG. They are intended as plug-in replacement for RAG systems with x5 faster inference
A Hub for Human-Centric 3D Vision
Our SPLADE models
-
naver/splade-cocondenser-ensembledistil
Feature Extraction • Updated • 504k • 60 -
naver/splade-v3
Feature Extraction • Updated • 86.7k • • 87 -
naver/efficient-splade-VI-BT-large-query
Fill-Mask • Updated • 264k • • 18 -
SPLADE-v3: New baselines for SPLADE
Paper • 2403.06789 • Published • 5
Provence: lightweight context pruning models for retrieval-augmented generation. Blogpost: https://huggingface.co/blog/nadiinchi/provence
Compression Models for RAG
-
Context Embeddings for Efficient Answer Generation in RAG
Paper • 2407.09252 • Published • 2 -
naver/cocom-v1-128-mistral-7b
Feature Extraction • 7B • Updated • 14 • 6 -
naver/cocom-v1-16-mistral-7b
Feature Extraction • 7B • Updated • 39 -
naver/cocom-v1-4-mistral-7b
Feature Extraction • 7B • Updated • 1 • 3
Multilingual Distilwhisper allows for better ASR performance in target languages by adding lightweight CLSR modules on top of whisper-small.
DUSt3R/MASt3R models
-
naver/MASt3R_ViTLarge_BaseDecoder_512_catmlpdpt_metric
Image-to-3D • 0.7B • Updated • 38.7k • 17 -
naver/DUSt3R_ViTLarge_BaseDecoder_512_dpt
Image-to-3D • 0.6B • Updated • 9.73k • 17 -
naver/DUSt3R_ViTLarge_BaseDecoder_512_linear
Image-to-3D • Updated • 177 -
naver/DUSt3R_ViTLarge_BaseDecoder_224_linear
Image-to-3D • Updated • 861
Online soft compression models for RAG. We release the models with llama-1B as compressor.
Provence: lightweight context pruning models for retrieval-augmented generation. Blogpost: https://huggingface.co/blog/nadiinchi/provence
PISCO models are compression models for RAG. They are intended as plug-in replacement for RAG systems with x5 faster inference
Compression Models for RAG
-
Context Embeddings for Efficient Answer Generation in RAG
Paper • 2407.09252 • Published • 2 -
naver/cocom-v1-128-mistral-7b
Feature Extraction • 7B • Updated • 14 • 6 -
naver/cocom-v1-16-mistral-7b
Feature Extraction • 7B • Updated • 39 -
naver/cocom-v1-4-mistral-7b
Feature Extraction • 7B • Updated • 1 • 3
A Hub for Human-Centric 3D Vision
Multilingual Distilwhisper allows for better ASR performance in target languages by adding lightweight CLSR modules on top of whisper-small.
Our SPLADE models
-
naver/splade-cocondenser-ensembledistil
Feature Extraction • Updated • 504k • 60 -
naver/splade-v3
Feature Extraction • Updated • 86.7k • • 87 -
naver/efficient-splade-VI-BT-large-query
Fill-Mask • Updated • 264k • • 18 -
SPLADE-v3: New baselines for SPLADE
Paper • 2403.06789 • Published • 5
DUSt3R/MASt3R models
-
naver/MASt3R_ViTLarge_BaseDecoder_512_catmlpdpt_metric
Image-to-3D • 0.7B • Updated • 38.7k • 17 -
naver/DUSt3R_ViTLarge_BaseDecoder_512_dpt
Image-to-3D • 0.6B • Updated • 9.73k • 17 -
naver/DUSt3R_ViTLarge_BaseDecoder_512_linear
Image-to-3D • Updated • 177 -
naver/DUSt3R_ViTLarge_BaseDecoder_224_linear
Image-to-3D • Updated • 861