MMTEB: Massive Multilingual Text Embedding Benchmark
Paper
•
2502.13595
•
Published
•
23
Open-Source AI Meetup
community
AI & ML interests
Open science and open source
Recent Activity
View all activity
SFEvent's activity
tomaarsen
authored
a
paper
1 day ago
Post
473
🚀 Try Out RAG Demo! 🚀
A Hugging Face Space where you can compare DeepSeek-R1 vs Llama-3 using Stuff RAG (Retrieval-Augmented Generation)!
🔍 Upload a PDF, ask questions, and see how both models perform in real-time!
Try out now:
zamal/Deepseek-R1-vs-LLama3
A Hugging Face Space where you can compare DeepSeek-R1 vs Llama-3 using Stuff RAG (Retrieval-Augmented Generation)!
🔍 Upload a PDF, ask questions, and see how both models perform in real-time!
Try out now:
zamal/Deepseek-R1-vs-LLama3
- 1 reply
Post
1947
I just released Sentence Transformers v3.4.0, featuring a memory leak fix, compatibility between the powerful Cached... losses and the Matryoshka loss modifier, and a bunch of fixes & small features.
🪆 Matryoshka & Cached loss compatibility
It is now possible to combine the powerful Cached... losses (which use in-batch negatives & a caching mechanism to allow for endless batch size & negatives) with the Matryoshka loss modifier which modifies a base loss such that it is trained not only on the maximum dimensionality (e.g. 1024 dimensions), but also on many lower dimensions (e.g. 768, 512, 256, 128, 64, 32).
After training, these models' embeddings can be truncated for faster retrieval, etc.
🎞️ Resolve memory leak when Model and Trainer are reinitialized
Due to a circular dependency between Trainer -> Model -> ModelCardData -> Trainer, deleting both the trainer & model still didn't free up the memory.
This led to a memory leak in scripts where you repeatedly do so.
➕ New Features
Many new small features, e.g. multi-GPU support for 'mine_hard_negatives', a 'margin' parameter to TripletEvaluator, and Matthews Correlation Coefficient in the BinaryClassificationEvaluator.
🐛 Bug Fixes
Also a bunch of fixes, for example that subsequent batches were not sorted when using the "no_duplicates" batch sampler. See the release notes for more details.
Full release notes: https://github.com/UKPLab/sentence-transformers/releases/tag/v3.4.0
Big thanks to all community members who assisted in this release. 10 folks with their first contribution this time around!
🪆 Matryoshka & Cached loss compatibility
It is now possible to combine the powerful Cached... losses (which use in-batch negatives & a caching mechanism to allow for endless batch size & negatives) with the Matryoshka loss modifier which modifies a base loss such that it is trained not only on the maximum dimensionality (e.g. 1024 dimensions), but also on many lower dimensions (e.g. 768, 512, 256, 128, 64, 32).
After training, these models' embeddings can be truncated for faster retrieval, etc.
🎞️ Resolve memory leak when Model and Trainer are reinitialized
Due to a circular dependency between Trainer -> Model -> ModelCardData -> Trainer, deleting both the trainer & model still didn't free up the memory.
This led to a memory leak in scripts where you repeatedly do so.
➕ New Features
Many new small features, e.g. multi-GPU support for 'mine_hard_negatives', a 'margin' parameter to TripletEvaluator, and Matthews Correlation Coefficient in the BinaryClassificationEvaluator.
🐛 Bug Fixes
Also a bunch of fixes, for example that subsequent batches were not sorted when using the "no_duplicates" batch sampler. See the release notes for more details.
Full release notes: https://github.com/UKPLab/sentence-transformers/releases/tag/v3.4.0
Big thanks to all community members who assisted in this release. 10 folks with their first contribution this time around!
Post
1441
zamal/Multimodal-Chat-PDF
🚀 Introducing Chat PDF Multimodal 💬
Interact with your PDF documents like never before! 🤯
Extract text & images, then ask context-aware questions based on both. Powered by RAG techniques & multimodal LLMs. Perfect for studying, research & more! 📝👀
Try it out now!!!! ✍️
#LlavaNext #MultimodalAI #Transformers
🚀 Introducing Chat PDF Multimodal 💬
Interact with your PDF documents like never before! 🤯
Extract text & images, then ask context-aware questions based on both. Powered by RAG techniques & multimodal LLMs. Perfect for studying, research & more! 📝👀
Try it out now!!!! ✍️
#LlavaNext #MultimodalAI #Transformers
Post
490
New state-of-the-art BERT-size retrieval model: *cde-small-v2* 🥳🍾
Hi everyone! We at Cornell are releasing a new retrieval model this week. It uses the contextual embeddings framework, is based on ModernBERT backbone, and gets state-of-the-art results on the MTEB benchmark for its model size (140M parameters). cde-small-v2 gets an average score of 65.6 across the 56 datasets and sees improvements from our previous model in *every* task domain (retrieval, classification, etc.).
We made a lot of changes to make this model work. First of all, ModernBERT has a better tokenizer, which probably helped this work out-of-the-box. We also followed the principles from the CDE paper and used harder clusters and better hard-negative filtering, which showed a small performance improvement. And we made a few small changes that have been shown to work on the larger models: we disabled weight decay, masked out the prefix tokens during pooling, and added a residual connection from the first-stage to the second-stage for better gradient flow.
We're still looking for a computer sponsor to help us scale CDE to larger models. Since it's now state-of-the-art at the 100M parameter scale, it seems to be a reasonable bet that we could train a state-of-the-art large model if we had the GPUs. If you're interested in helping with this, please reach out!
Here's a link to the model: jxm/cde-small-v2
And here's a link to the paper: Contextual Document Embeddings (2410.02525)
Hi everyone! We at Cornell are releasing a new retrieval model this week. It uses the contextual embeddings framework, is based on ModernBERT backbone, and gets state-of-the-art results on the MTEB benchmark for its model size (140M parameters). cde-small-v2 gets an average score of 65.6 across the 56 datasets and sees improvements from our previous model in *every* task domain (retrieval, classification, etc.).
We made a lot of changes to make this model work. First of all, ModernBERT has a better tokenizer, which probably helped this work out-of-the-box. We also followed the principles from the CDE paper and used harder clusters and better hard-negative filtering, which showed a small performance improvement. And we made a few small changes that have been shown to work on the larger models: we disabled weight decay, masked out the prefix tokens during pooling, and added a residual connection from the first-stage to the second-stage for better gradient flow.
We're still looking for a computer sponsor to help us scale CDE to larger models. Since it's now state-of-the-art at the 100M parameter scale, it seems to be a reasonable bet that we could train a state-of-the-art large model if we had the GPUs. If you're interested in helping with this, please reach out!
Here's a link to the model: jxm/cde-small-v2
And here's a link to the paper: Contextual Document Embeddings (2410.02525)
Post
4610
🏎️ Today I'm introducing a method to train static embedding models that run 100x to 400x faster on CPU than common embedding models, while retaining 85%+ of the quality! Including 2 fully open models: training scripts, datasets, metrics.
We apply our recipe to train 2 Static Embedding models that we release today! We release:
2️⃣ an English Retrieval model and a general-purpose Multilingual similarity model (e.g. classification, clustering, etc.), both Apache 2.0
🧠 my modern training strategy: ideation -> dataset choice -> implementation -> evaluation
📜 my training scripts, using the Sentence Transformers library
📊 my Weights & Biases reports with losses & metrics
📕 my list of 30 training and 13 evaluation datasets
The 2 Static Embedding models have the following properties:
🏎️ Extremely fast, e.g. 107500 sentences per second on a consumer CPU, compared to 270 for 'all-mpnet-base-v2' and 56 for 'gte-large-en-v1.5'
0️⃣ Zero active parameters: No Transformer blocks, no attention, not even a matrix multiplication. Super speed!
📏 No maximum sequence length! Embed texts at any length (note: longer texts may embed worse)
📐 Linear instead of exponential complexity: 2x longer text takes 2x longer, instead of 2.5x or more.
🪆 Matryoshka support: allow you to truncate embeddings with minimal performance loss (e.g. 4x smaller with a 0.56% perf. decrease for English Similarity tasks)
Check out the full blogpost if you'd like to 1) use these lightning-fast models or 2) learn how to train them with consumer-level hardware: https://huggingface.co/blog/static-embeddings
The blogpost contains a lengthy list of possible advancements; I'm very confident that our 2 models are only the tip of the iceberg, and we may be able to get even better performance.
Alternatively, check out the models:
* sentence-transformers/static-retrieval-mrl-en-v1
* sentence-transformers/static-similarity-mrl-multilingual-v1
We apply our recipe to train 2 Static Embedding models that we release today! We release:
2️⃣ an English Retrieval model and a general-purpose Multilingual similarity model (e.g. classification, clustering, etc.), both Apache 2.0
🧠 my modern training strategy: ideation -> dataset choice -> implementation -> evaluation
📜 my training scripts, using the Sentence Transformers library
📊 my Weights & Biases reports with losses & metrics
📕 my list of 30 training and 13 evaluation datasets
The 2 Static Embedding models have the following properties:
🏎️ Extremely fast, e.g. 107500 sentences per second on a consumer CPU, compared to 270 for 'all-mpnet-base-v2' and 56 for 'gte-large-en-v1.5'
0️⃣ Zero active parameters: No Transformer blocks, no attention, not even a matrix multiplication. Super speed!
📏 No maximum sequence length! Embed texts at any length (note: longer texts may embed worse)
📐 Linear instead of exponential complexity: 2x longer text takes 2x longer, instead of 2.5x or more.
🪆 Matryoshka support: allow you to truncate embeddings with minimal performance loss (e.g. 4x smaller with a 0.56% perf. decrease for English Similarity tasks)
Check out the full blogpost if you'd like to 1) use these lightning-fast models or 2) learn how to train them with consumer-level hardware: https://huggingface.co/blog/static-embeddings
The blogpost contains a lengthy list of possible advancements; I'm very confident that our 2 models are only the tip of the iceberg, and we may be able to get even better performance.
Alternatively, check out the models:
* sentence-transformers/static-retrieval-mrl-en-v1
* sentence-transformers/static-similarity-mrl-multilingual-v1
- 1 reply
Sri-Vigneshwar-DJ
posted
an
update
about 1 month ago
Post
687
Checkout phi-4 from Microsoft, dropped a day ago... If you ❤️ the Phi series, then here is the GGUF -
Sri-Vigneshwar-DJ/phi-4-GGUF. phi-4 is a 14B highly efficient open LLM that beats much larger models at math and reasoning - check out evaluations on the Open LLM.
Technical paper - https://arxiv.org/pdf/2412.08905 ; The Data Synthesis approach is interesting
Technical paper - https://arxiv.org/pdf/2412.08905 ; The Data Synthesis approach is interesting
ethanhe
authored
a
paper
about 2 months ago
Sri-Vigneshwar-DJ
posted
an
update
about 2 months ago
Post
2080
Just sharing a thought: I started using DeepSeek V3 a lot, and an idea struck me about agents "orchestrating during inference" on a test-time compute model like DeepSeek V3 or the O1 series.
Agents (Instruction + Function Calls + Memory) execute during inference, and based on the output decision, a decision is made to scale the time to reason or perform other tasks.
Agents (Instruction + Function Calls + Memory) execute during inference, and based on the output decision, a decision is made to scale the time to reason or perform other tasks.
Sri-Vigneshwar-DJ
posted
an
update
about 2 months ago
Post
2348
Combining smolagents with Anthropic’s best practices simplifies building powerful AI agents:
1. Code-Based Agents: Write actions as Python code, reducing steps by 30%.
2. Prompt Chaining: Break tasks into sequential subtasks with validation gates.
3. Routing: Classify inputs and direct them to specialized handlers.
4. Fallback: Handle tasks even if classification fails.
https://huggingface.co/blog/Sri-Vigneshwar-DJ/building-effective-agents-with-anthropics-best-pra
1. Code-Based Agents: Write actions as Python code, reducing steps by 30%.
2. Prompt Chaining: Break tasks into sequential subtasks with validation gates.
3. Routing: Classify inputs and direct them to specialized handlers.
4. Fallback: Handle tasks even if classification fails.
https://huggingface.co/blog/Sri-Vigneshwar-DJ/building-effective-agents-with-anthropics-best-pra
Post
3010
That didn't take long! Nomic AI has finetuned the new ModernBERT-base encoder model into a strong embedding model for search, classification, clustering and more!
Details:
🤖 Based on ModernBERT-base with 149M parameters.
📊 Outperforms both nomic-embed-text-v1 and nomic-embed-text-v1.5 on MTEB!
🏎️ Immediate FA2 and unpacking support for super efficient inference.
🪆 Trained with Matryoshka support, i.e. 2 valid output dimensionalities: 768 and 256.
➡️ Maximum sequence length of 8192 tokens!
2️⃣ Trained in 2 stages: unsupervised contrastive data -> high quality labeled datasets.
➕ Integrated in Sentence Transformers, Transformers, LangChain, LlamaIndex, Haystack, etc.
🏛️ Apache 2.0 licensed: fully commercially permissible
Try it out here: nomic-ai/modernbert-embed-base
Very nice work by Zach Nussbaum and colleagues at Nomic AI.
Details:
🤖 Based on ModernBERT-base with 149M parameters.
📊 Outperforms both nomic-embed-text-v1 and nomic-embed-text-v1.5 on MTEB!
🏎️ Immediate FA2 and unpacking support for super efficient inference.
🪆 Trained with Matryoshka support, i.e. 2 valid output dimensionalities: 768 and 256.
➡️ Maximum sequence length of 8192 tokens!
2️⃣ Trained in 2 stages: unsupervised contrastive data -> high quality labeled datasets.
➕ Integrated in Sentence Transformers, Transformers, LangChain, LlamaIndex, Haystack, etc.
🏛️ Apache 2.0 licensed: fully commercially permissible
Try it out here: nomic-ai/modernbert-embed-base
Very nice work by Zach Nussbaum and colleagues at Nomic AI.
tomaarsen
authored
a
paper
2 months ago
freddyaboulton
posted
an
update
2 months ago
Post
1708
Just created a Gradio space for playing with the new OAI realtime voice API!
freddyaboulton/openai-realtime-voice
freddyaboulton/openai-realtime-voice
freddyaboulton
posted
an
update
2 months ago
Post
932
Gemini can talk 🗣️
Check out the new multimodal API from Google on @akhaliq 's anychat or my space. It's very fast and smart 🍓
https://huggingface.co/spaces/freddyaboulton/gemini-voicehttps://huggingface.co/spaces/akhaliq/anychat
Check out the new multimodal API from Google on @akhaliq 's anychat or my space. It's very fast and smart 🍓
https://huggingface.co/spaces/freddyaboulton/gemini-voicehttps://huggingface.co/spaces/akhaliq/anychat
- 1 reply
freddyaboulton
posted
an
update
2 months ago
Post
2190
Version 0.0.21 of gradio-pdf now properly loads chinese characters!
freddyaboulton
posted
an
update
2 months ago
Post
1593
Hello Llama 3.2! 🗣️🦙
Build a Siri-like coding assistant that responds to "Hello Llama" in 100 lines of python! All with Gradio, webRTC 😎
freddyaboulton/hey-llama-code-editor
Build a Siri-like coding assistant that responds to "Hello Llama" in 100 lines of python! All with Gradio, webRTC 😎
freddyaboulton/hey-llama-code-editor
freddyaboulton
posted
an
update
2 months ago
Post
1140
Just created a cookbook of real time audio/video spaces created using Gradio and WebRTC ⚡️
Use this and the [docs](https://freddyaboulton.github.io/gradio-webrtc/) to get started building the next gen of AI apps!
freddyaboulton/gradio-webrtc-cookbook-6758ba7745aeca7b1be7de0f
Use this and the [docs](https://freddyaboulton.github.io/gradio-webrtc/) to get started building the next gen of AI apps!
freddyaboulton/gradio-webrtc-cookbook-6758ba7745aeca7b1be7de0f
-
- 2 replies
Post
1671
First Global and Dense Open Embedding Dataset of Earth! 🌍 🤗
Introducing the Major TOM embeddings dataset, created in collaboration with CloudFerro S.A. 🔶 and Φ-lab at the European Space Agency (ESA) 🛰️. Together with @mikonvergence and Jędrzej S. Bojanowski, we present the first open-access dataset of Copernicus embeddings, offering dense, global coverage across the full acquisition areas of Sentinel-1 and Sentinel-2 sensors.
💡 Highlights:
📊 Data: Over 8 million Sentinel-1 & Sentinel-2 images processed, distilling insights from 9.368 trillion pixels of raw data.
🧠 Models: Foundation models include SigLIP, DINOv2, and SSL4EO.
📦 Scale: 62 TB of raw satellite data processed into 170M+ embeddings.
This project delivers open and free vectorized expansions of Major-TOM/README datasets, setting a new standard for embedding releases and enabling lightweight, scalable ingestion of Earth Observation (EO) data for countless applications.
🤗 Explore the datasets:
Major-TOM/Core-S2L1C-SSL4EO
Major-TOM/Core-S1RTC-SSL4EO
Major-TOM/Core-S2RGB-DINOv2
Major-TOM/Core-S2RGB-SigLIP
📖 Check paper: Global and Dense Embeddings of Earth: Major TOM Floating in the Latent Space (2412.05600)
💻 Code notebook: https://github.com/ESA-PhiLab/Major-TOM/blob/main/05-Generate-Major-TOM-Embeddings.ipynb
Introducing the Major TOM embeddings dataset, created in collaboration with CloudFerro S.A. 🔶 and Φ-lab at the European Space Agency (ESA) 🛰️. Together with @mikonvergence and Jędrzej S. Bojanowski, we present the first open-access dataset of Copernicus embeddings, offering dense, global coverage across the full acquisition areas of Sentinel-1 and Sentinel-2 sensors.
💡 Highlights:
📊 Data: Over 8 million Sentinel-1 & Sentinel-2 images processed, distilling insights from 9.368 trillion pixels of raw data.
🧠 Models: Foundation models include SigLIP, DINOv2, and SSL4EO.
📦 Scale: 62 TB of raw satellite data processed into 170M+ embeddings.
This project delivers open and free vectorized expansions of Major-TOM/README datasets, setting a new standard for embedding releases and enabling lightweight, scalable ingestion of Earth Observation (EO) data for countless applications.
🤗 Explore the datasets:
Major-TOM/Core-S2L1C-SSL4EO
Major-TOM/Core-S1RTC-SSL4EO
Major-TOM/Core-S2RGB-DINOv2
Major-TOM/Core-S2RGB-SigLIP
📖 Check paper: Global and Dense Embeddings of Earth: Major TOM Floating in the Latent Space (2412.05600)
💻 Code notebook: https://github.com/ESA-PhiLab/Major-TOM/blob/main/05-Generate-Major-TOM-Embeddings.ipynb
- 1 reply
Vishnou
authored
a
paper
3 months ago
Post
5743
I just released Sentence Transformers v3.3.0 & it's huge! 4.5x speedup for CPU with OpenVINO int8 static quantization, training with prompts for a free perf. boost, PEFT integration, evaluation on NanoBEIR, and more! Details:
1. We integrate Post-Training Static Quantization using OpenVINO, a very efficient solution for CPUs that processes 4.78x as many texts per second on average, while only hurting performance by 0.36% on average. There's a new
2. We add the option to train with prompts, e.g. strings like "query: ", "search_document: " or "Represent this sentence for searching relevant passages: ". It's as simple as using the
3. Sentence Transformers now supports training PEFT adapters via 7 new methods for adding new adapters or loading pre-trained ones. You can also directly load a trained adapter with SentenceTransformer as if it's a normal model. Very useful for e.g. 1) training multiple adapters on 1 base model, 2) training bigger models than otherwise possible, or 3) cheaply hosting multiple models by switching multiple adapters on 1 base model.
4. We added easy evaluation on NanoBEIR, a subset of BEIR a.k.a. the MTEB Retrieval benchmark. It contains 13 datasets with 50 queries and up to 10k documents each. Evaluation is fast, and can easily be done during training to track your model's performance on general-purpose information retrieval tasks.
Additionally, we also deprecate Python 3.8, add better compatibility with Transformers v4.46.0, and more. Read the full release notes here: https://github.com/UKPLab/sentence-transformers/releases/tag/v3.3.0
1. We integrate Post-Training Static Quantization using OpenVINO, a very efficient solution for CPUs that processes 4.78x as many texts per second on average, while only hurting performance by 0.36% on average. There's a new
export_static_quantized_openvino_model method to quantize a model.2. We add the option to train with prompts, e.g. strings like "query: ", "search_document: " or "Represent this sentence for searching relevant passages: ". It's as simple as using the
prompts argument in SentenceTransformerTrainingArguments. Our experiments show that you can easily reach 0.66% to 0.90% relative performance improvement on NDCG@10 at no extra cost by adding "query: " before each training query and "document: " before each training answer.3. Sentence Transformers now supports training PEFT adapters via 7 new methods for adding new adapters or loading pre-trained ones. You can also directly load a trained adapter with SentenceTransformer as if it's a normal model. Very useful for e.g. 1) training multiple adapters on 1 base model, 2) training bigger models than otherwise possible, or 3) cheaply hosting multiple models by switching multiple adapters on 1 base model.
4. We added easy evaluation on NanoBEIR, a subset of BEIR a.k.a. the MTEB Retrieval benchmark. It contains 13 datasets with 50 queries and up to 10k documents each. Evaluation is fast, and can easily be done during training to track your model's performance on general-purpose information retrieval tasks.
Additionally, we also deprecate Python 3.8, add better compatibility with Transformers v4.46.0, and more. Read the full release notes here: https://github.com/UKPLab/sentence-transformers/releases/tag/v3.3.0