SFEvent (Open-Source AI Meetup)

tomaarsen

posted an update about 2 months ago

Post

3390

I just released Sentence Transformers v4.1; featuring ONNX and OpenVINO backends for rerankers offering 2-3x speedups and improved hard negatives mining which helps prepare stronger training datasets. Details:

🏎️ ONNX, OpenVINO, Optimization, Quantization
- I've added ONNX and OpenVINO support with just one extra argument: "backend" when loading the CrossEncoder reranker, e.g.: CrossEncoder("cross-encoder/ms-marco-MiniLM-L6-v2", backend="onnx")
- The export_optimized_onnx_model, export_dynamic_quantized_onnx_model, and export_static_quantized_openvino_model functions now work with CrossEncoder rerankers, allowing you to optimize (e.g. fusions, gelu approximations, etc.) or quantize (int8 weights) rerankers.
- I've uploaded ~340 ONNX & OpenVINO models for all existing models under the cross-encoder Hugging Face organization. You can use these without having to export when loading.

⛏ Improved Hard Negatives Mining
- Added 'absolute_margin' and 'relative_margin' arguments to mine_hard_negatives.
- absolute_margin ensures that sim(query, negative) < sim(query, positive) - absolute_margin, i.e. an absolute margin between the negative & positive similarities.
- relative_margin ensures that sim(query, negative) < sim(query, positive) * (1 - relative_margin), i.e. a relative margin between the negative & positive similarities.
- Inspired by the excellent NV-Retriever paper from NVIDIA.

And several other small improvements. Check out the full release notes here: https://github.com/UKPLab/sentence-transformers/releases/tag/v4.1.0

With this release, I introduce near-feature parity between the SentenceTransformer embedding & CrossEncoder reranker models, which I've wanted to do for quite some time! With rerankers very strongly supported now, it's time to look forward to other useful architectures!

rajistics

posted an update about 2 months ago

Post

3472

Having some fun with long context benchmarks (watch the video!!)

NoLiMA: NoLiMa: Long-Context Evaluation Beyond Literal Matching (2502.05167)
Fiction LiveBench: https://fiction.live/stories/Fiction-liveBench-Mar-25-2025/oQdzQvKHw8JyXbN87
Michalenglo: https://deepmind.google/research/publications/117639/
LongGenBench: Spinning the Golden Thread: Benchmarking Long-Form Generation in Language Models (2409.02076)
NeedleBench: NeedleBench: Can LLMs Do Retrieval and Reasoning in 1 Million Context Window? (2407.11963)
RULER: RULER: What's the Real Context Size of Your Long-Context Language Models? (2404.06654)

For more: https://www.reddit.com/r/rajistics/comments/1jxwk29/long_context_llm_benchmarks_video/

let me know if you like these posts

lianghsun

posted an update about 2 months ago

Post

2525

With the arrival of Twinkle April — Twinkle AI’s annual open-source celebration held every April — our community is excited to unveil its very first project:

📊 Twinkle Eval (https://github.com/ai-twinkle/Eval), a next-generation evaluation tool led by our contributor @tedslin .

Unlike traditional evaluation tools like iKala’s ievals (https://github.com/ikala-ai/ievals), which can only evaluate language models (LMs) one sample at a time, Twinkle Eval is designed with Large Reasoning Models (LRMs) in mind. As reasoning time increases with more complex models, traditional tools become increasingly inefficient 😲 — for example, evaluating LRMs on the ikala/tmmluplus benchmark could take *
half a day without finishing.

One question we were especially curious about:
Does shuffling multiple-choice answer order impact model accuracy? 🤔
→ See: "Change Answer Order Can Decrease MMLU Accuracy" – arXiv:2406.19470v1

To address these challenges, Twinkle Eval brings three key innovations to the table:

1️⃣ Parallelized evaluation of samples
2️⃣ Multi-round testing for stability
3️⃣ Randomized answer order to test robustness

After running experiments, we observed that Twinkle Eval can speed up evaluation by up to 15× 🚀🚀. Interestingly, most models scored slightly lower under the 2️⃣3️⃣ test settings compared to their claimed performance — suggesting further benchmarking is needed.

This framework also comes with additional tunable parameters and detailed logging of LM behavior per question — perfect for those who want to dive deeper. 😆

If you find Twinkle Eval useful, please ⭐ the project and help spread the word 🤗

5 replies

·

zamal

posted an update about 2 months ago

Post

1946

🚀 DeepGit Lite is live! 🔍✨

Hey folks!
Just launched DeepGit Lite — a lighter version of DeepGit with fewer components under the hood.
It won’t perform quite like the full powerhouse, but it’s great for a quick peek and first-hand feel! ⚙️👀

Give it a spin and tell us what you think!
👉 Try it here zamal/DeepGit-lite
#opensource #DeepGit #gradio #githubresearch

3 replies

·

zamal

posted an update 2 months ago

Post

2587

DeepGit: Your GitHub Gold Digger! 💰🚀
Hey Hugging Face gang! Meet DeepGit—my open-source sidekick that rips through GitHub to snag repos that fit you. Done with dead-end searches? Me too. Built it with LangGraph and some dope tricks:
Embeddings grab the good stuff (HF magic, baby!)

Re-ranking nails the best picks

Snoops docs, code, and buzz in one slick flow

Drops a clean list of hidden gems 💎

Unearth that sneaky ML lib or Python gem—run python app.py or langgraph dev and boom! Peek it at https://github.com/zamalali/DeepGit. Fork it, tweak it, love it—Docker’s in, HF vibes are strong. Drop a 🌟 or a crazy idea—I’m pumped to jam with you all! 🪂

tomaarsen

posted an update 2 months ago

Post

2583

‼️Sentence Transformers v4.0 is out! You can now train and finetune reranker models with multi-GPU training, bf16 support, loss logging, callbacks & much more. I also prove that finetuning on your domain helps much more than you might think.

1️⃣ Reranker Training Refactor
Reranker models can now be trained using an extensive trainer with a lot of powerful features:
- MultiGPU Training (Data Parallelism (DP) and Distributed Data Parallelism (DDP))
- bf16 training support; loss logging
- Evaluation datasets + evaluation loss
- Improved callback support + an excellent Weights & Biases integration
- Gradient checkpointing, gradient accumulation
- Model card generation
- Resuming from a training checkpoint without performance loss
- Hyperparameter Optimization
and much more!

Read my detailed blogpost to learn about the components that make up this new training approach: https://huggingface.co/blog/train-reranker
Notably, the release is fully backwards compatible: all deprecations are soft, meaning that they still work but emit a warning informing you how to upgrade.

2️⃣ New Reranker Losses
- 11 new losses:
- 2 traditional losses: BinaryCrossEntropy and CrossEntropy
- 2 distillation losses: MSE and MarginMSE
- 2 in-batch negatives losses: MNRL (a.k.a. InfoNCE) and CMNRL
- 5 learning to rank losses: Lambda, p-ListMLE, ListNet, RankNet, ListMLE

3️⃣ New Reranker Documentation
- New Training Overview, Loss Overview, API Reference docs
- 5 new, 1 refactored training examples docs pages
- 13 new, 6 refactored training scripts
- Migration guides (2.x -> 3.x, 3.x -> 4.x)

4️⃣ Blogpost
Alongside the release, I've written a blogpost where I finetune ModernBERT on a generic question-answer dataset. My finetunes easily outperform all general-purpose reranker models, even models 4x as big. Finetuning on your domain is definitely worth it: https://huggingface.co/blog/train-reranker

See the full release notes here: https://github.com/UKPLab/sentence-transformers/releases/v4.0.1

louisbrulenaudet

posted an update 2 months ago

Post

1078

I’ve just released logfire-callback on PyPI, designed to facilitate monitoring of Hugging Face Transformer training loops using Pydantic Logfire 🤗

The callback will automatically log training start with configuration parameters, periodic metrics and training completion ⏱️

Install the package using pip:

pip install logfire-callback

First, ensure you have a Logfire API token and set it as an environment variable:

export LOGFIRE_TOKEN=your_logfire_token

Then use the callback in your training code:

from transformers import Trainer, TrainingArguments
from logfire_callback import LogfireCallback

# Initialize your model, dataset, etc.

training_args = TrainingArguments(
    output_dir="./results",
    num_train_epochs=3,
    # ... other training arguments
)

trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=train_dataset,
    callbacks=[LogfireCallback()]  # Add the Logfire callback here
)

trainer.train()

If you have any feedback, please reach out at @louisbrulenaudet

tomaarsen

posted an update 3 months ago

Post

6768

An assembly of 18 European companies, labs, and universities have banded together to launch 🇪🇺 EuroBERT! It's a state-of-the-art multilingual encoder for 15 European languages, designed to be finetuned for retrieval, classification, etc.

🇪🇺 15 Languages: English, French, German, Spanish, Chinese, Italian, Russian, Polish, Portuguese, Japanese, Vietnamese, Dutch, Arabic, Turkish, Hindi
3️⃣ 3 model sizes: 210M, 610M, and 2.1B parameters - very very useful sizes in my opinion
➡️ Sequence length of 8192 tokens! Nice to see these higher sequence lengths for encoders becoming more common.
⚙️ Architecture based on Llama, but with bi-directional (non-causal) attention to turn it into an encoder. Flash Attention 2 is supported.
🔥 A new Pareto frontier (stronger *and* smaller) for multilingual encoder models
📊 Evaluated against mDeBERTa, mGTE, XLM-RoBERTa for Retrieval, Classification, and Regression (after finetuning for each task separately): EuroBERT punches way above its weight.
📝 Detailed paper with all details, incl. data: FineWeb for English and CulturaX for multilingual data, The Stack v2 and Proof-Pile-2 for code.

Check out the release blogpost here: https://huggingface.co/blog/EuroBERT/release
* EuroBERT/EuroBERT-210m
* EuroBERT/EuroBERT-610m
* EuroBERT/EuroBERT-2.1B

The next step is for researchers to build upon the 3 EuroBERT base models and publish strong retrieval, zero-shot classification, etc. models for all to use. I'm very much looking forward to it!

1 reply

·

zamal

posted an update 3 months ago

Post

2027

🚀 ftBoost is LIVE – Stop Struggling with Fine-Tuning Data!

Alright folks, if you’re tired of manually crafting fine-tuning datasets, ftBoost is here to do the heavy lifting. One-click, LangChain-Groq-powered data augmentation that scales your training data in OpenAI, Gemini, Mistral, and LLaMA formats—automatically.

🔥 What’s inside?
✅ Smart Augmentations – Paraphrasing, back translation, synonym swapping & synthetic noise.
✅ No more JSONL headaches – Auto-formats everything for OpenAI, Gemini, Mistral & LLaMA.
✅ Custom tuning – Adjust similarity, diversity, and fluency in real-time.
✅ Upload, generate, download – That’s it.

⚡ If you’re fine-tuning LLMs, this will save you hours.

🚀 Try it now: 👉 zamal/Finetune-Boost

🌟 Give us a star on GitHub!

Let me know what you think & how it boosts your workflow! 🔥

tomaarsen

authored a paper 3 months ago

MMTEB: Massive Multilingual Text Embedding Benchmark

Paper • 2502.13595 • Published Feb 19 • 34

louisbrulenaudet

posted an update 3 months ago

Post

3414

I am pleased to introduce my first project built upon Hugging Face’s smolagents framework, integrated with Alpaca for financial market analysis automation 🦙🤗

The project implements technical indicators such as the Relative Strength Index (RSI) and Bollinger Bands to provide momentum and volatility analysis. Market data is retrieved through the Alpaca API, enabling access to historical price information across various timeframes.

AI-powered insights are generated using Hugging Face’s inference API, facilitating the analysis of market trends through natural language processing with DuckDuckGo search integration for real-time sentiment analysis based on financial news 🦆

Link to the GitHub project: https://github.com/louisbrulenaudet/agentic-market-tool

zamal

posted an update 4 months ago

Post

623

🚀 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

1 reply

·

tomaarsen

posted an update 4 months ago

Post

2316

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!

zamal

posted an update 4 months ago

Post

1496

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

tomaarsen

posted an update 5 months ago

Post

4756

🏎️ 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

1 reply

·

lianghsun

posted an update 5 months ago

Post

2828

🖖 Let me introduce the work I've done over the past three months: 𝗟𝗹𝗮𝗺𝗮-𝟯.𝟮-𝗧𝗮𝗶𝘄𝗮𝗻-𝟯𝗕 and 𝗟𝗹𝗮𝗺𝗮-𝟯.𝟮-𝗧𝗮𝗶𝘄𝗮𝗻-𝟯𝗕-𝗜𝗻𝘀𝘁𝗿𝘂𝗰𝘁, now open-sourced on 🤗 Hugging Face.

𝗹𝗶𝗮𝗻𝗴𝗵𝘀𝘂𝗻/𝗟𝗹𝗮𝗺𝗮-𝟯.𝟮-𝗧𝗮𝗶𝘄𝗮𝗻-𝟯𝗕: This model is built on top of 𝗺𝗲𝘁𝗮-𝗹𝗹𝗮𝗺𝗮/𝗟𝗹𝗮𝗺𝗮-𝟯.𝟮-𝟯𝗕 with continual pretraining. The training dataset consists of a mixture of Traditional Chinese and multilingual texts in specific proportions, including 20B tokens of Traditional Chinese text.

𝗹𝗶𝗮𝗻𝗴𝗵𝘀𝘂𝗻/𝗟𝗹𝗮𝗺𝗮-𝟯.𝟮-𝗧𝗮𝗶𝘄𝗮𝗻-𝟯𝗕-𝗜𝗻𝘀𝘁𝗿𝘂𝗰𝘁: This is a fine-tuned conversational model based on the foundation model.

This Llama-3.2-Taiwan open-source project is currently a one-person effort (yes, I did everything from text preparation — so exhausting!). If you're interested, feel free to join the Discord server for discussions.

🅱🅴🅽🅲🅷🅼🅰🆁🅺🅸🅽🅶

The evaluation was conducted using ikala/tmmluplus, though the README page does not yet reflect the latest results. The performance is close to the previous versions, indicating that further improvements might require adding more specialized knowledge in the datasets.

🅰 🅲🅰🅻🅻 🅵🅾🆁 🆂🆄🅿🅿🅾🆁🆃

If anyone is willing to provide compute resources, it would be greatly appreciated to help this project continue and grow. 💪

---
🏔️ Foundation model: lianghsun/Llama-3.2-Taiwan-3B
🤖 Instruction model: lianghsun/Llama-3.2-Taiwan-3B-Instruct
⚡ GGUF: lianghsun/Llama-3.2-Taiwan-3B-Instruct-GGUF

4 replies

·

Sri-Vigneshwar-DJ

posted an update 5 months ago

Post

768

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

Sri-Vigneshwar-DJ

posted an update 5 months ago

Post

2091

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.

Sri-Vigneshwar-DJ

posted an update 5 months ago

Post

2353

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

tomaarsen

posted an update 5 months ago

Post

3084

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.

Open-Source AI Meetup

AI & ML interests

Recent Activity

SFEvent's activity

MMTEB: Massive Multilingual Text Embedding Benchmark

AI & ML interests

Recent Activity

Team members 571

SFEvent's activity