You can now use the "Synthetic Data Generator" at a much larger scale with your preferred inference engine: Ollama, vLLM, TGI, and serverless inference! 🔥
We've added a new chapter about the very basics of Argilla to the Hugging Face NLP course. Learn how to set up an Argilla instance, load & annotate datasets, and export them to the Hub.
📝ChemQwen-vL is a vision-language model fine-tuned based on the Qwen2VL-2B Instruct model. It has been trained using the International Chemical Identifier (InChI) format for chemical compounds and is optimized for chemical compound identification. The model excels at generating the InChI and providing descriptions of chemical compounds based on their images. Its architecture operates within a multi-modal framework, combining image-text-text capabilities. It has been fine-tuned using datasets from: https://iupac.org/projects/
@meg, one of the best researchers in AI ethics, makes a critical point about autonomy: fully autonomous systems carry unknowable risks because they operate on computer logic rather than human logic.
The solution? Build systems that support & assist rather than override human decisions.
I highly recommend reading the blog post written by Meg, @evijit@sasha and @giadap. They define different levels of agent autonomy & provide a values-based analysis of risks, benefits, and uses of AI agents to help you make better decisions.
🔥 The AI Agent hype is real! This blog post deep dives into everything you need to know before deploying them: from key definitions to practical recommendations. A must-read for anyone building the future of autonomous systems.
📊 Key insight: A clear table breaking down the 5 levels of AI agents - from simple processors to fully autonomous systems. Essential framework for understanding where your agent stands on the autonomy spectrum
⚖️ Deep analysis of 15 core values reveals critical trade-offs: accuracy, privacy, safety, equity & more. The same features that make agents powerful can make them risky. Understanding these trade-offs is crucial for responsible deployment
🎯 6 key recommendations for the road ahead: - Create rigorous evaluation protocols - Study societal effects - Understand ripple effects - Improve transparency - Open source can make a positive difference - Monitor base model evolution
FineWeb2 is a massive multilingual dataset for pre-training language models. Like any web-scale dataset, it contains low-quality content. How can we improve it?
Over the past months, an amazing community of 400+ annotators has been labelling content quality (using Argilla) across 23 languages through the FineWeb-C initiative.
Today, I'm happy to share the first classifier trained on this data.
🔍 What we've built:
- A lightweight classifier that efficiently removes low-quality content - 90%+ precision demonstrated on Danish & Swedish - Can process the 43M+ documents in Danish FineWeb2 with minimal compute
🌍 Why this matters: The approach can be reproduced for any of the 23 languages in FineWeb-C (data-is-better-together/fineweb-c). We can improve training data quality at scale without massive compute resources by starting with community annotations and training small, efficient classifiers.
This week a few more languages have got 1,000 annotations for the educational quality of data from HuggingFaceFW/fineweb-2.
Why should you care?
The quality of pre-training data can have a big impact on the performance of downstream language models trained on that data (HuggingFaceFW/blogpost-fineweb-v1).
Being able to filter by educational quality is on way of improving the quality of the data you use for training an LLM. Very importantly this approach can also reduce the amount of data needed for pertaining.
Why not use an LLM?
LLMs can be used to annotate educational quality for a subset of data. This data can then be used to train a smaller encoder only model to label the full dataset. However, this may not work well for languages outside of english. This is where fineweb-c (community) comes in.
The community is annotating the educational quality of fineweb2 data. Currently 114 languages have some annotations. These annotations will enable a number of things:
- Evaluate whether an LLM can label the educational quality for texts in that language well - Directly be used for training quality classifiers - Help discover other rules and huerisitcs for refining fineweb2 further for different languages.
❤️🔥Stranger Zone's MidJourney Mix Model Adapter is trending on the Very Model Page, with over 45,000+ downloads. Additionally, the Super Realism Model Adapter has over 52,000+ downloads, remains the top two adapter on Stranger Zone! strangerzonehf/Flux-Midjourney-Mix2-LoRA, strangerzonehf/Flux-Super-Realism-LoRA
3C3H AraGen Leaderboard welcomes today deepseek-ai/DeepSeek-V3 and 12 other models (including the late gpt-3.5 💀) to the ranking of best LLMs in Arabic !
Observations: - DeepSeek-v3 ranked 3rd and only Open model among the top 5 !
- A 14B open model (Qwen/Qwen2.5-14B-Instruct) outperforms gpt-3.5-turbo-0125 (from last year). This shows how much we came in advancing and supporting Arabic presence within the LLM ecosystem !
- Contrary to what observed in likelihood-acc leaderboards (like OALL/Open-Arabic-LLM-Leaderboard) further finetuned models like maldv/Qwentile2.5-32B-Instruct actually decreased the performance compared to the original model Qwen/Qwen2.5-32B-Instruct. It's worth to note that the decrease is statiscally insignificant which imply that at best, the out-domain finetuning do not really hurts the model original capabilities acquired during pretraining. Previous work addressed this (finetuning VS pretraining) but more investigation in this regard is required (any PhDs here ? This could be your question ...)
🎯Fine-tuning SmolLM2 on a lightweight synthetic reasoning dataset for reasoning-specific tasks. Future updates will focus on lightweight, blazing-fast reasoning models. Until then, check out the blog for fine-tuning details.
🎯Triangulum is a collection of pretrained and instruction-tuned generative models, designed for multilingual applications. These models are trained using synthetic datasets based on long chains of thought, enabling them to perform complex reasoning tasks effectively.
~75% on the challenging GPQA with only 40M parameters 🔥🥳
GREAT ACHIEVEMENT ! Or is it ?
This new Work, "Data Laundering: Artificially Boosting Benchmark Results through Knowledge Distillation", take out the mystery about many models i personally suspected their results. Speacially on leaderboards other than the english one, Like the Open Arabic LLM Leaderbaord OALL/Open-Arabic-LLM-Leaderboard.
The authors of this work, first started by training a model on the GPQA data, which, unsurprisingly, led to the model achieving 100% performance.
Afterward, they trained what they referred to as a 'legitimate' model on legitimate data (MedMCQA). However, they introduced a distillation loss from the earlier, 'cheated' model.
What they discovered was fascinating: the knowledge of GPQA leaked through this distillation loss, even though the legitimate model was never explicitly trained on GPQA during this stage.
This raises important questions about the careful use of distillation in model training, especially when the training data is opaque. As they demonstrated, it’s apparently possible to (intentionally or unintentionally) leak test data through this method.
🔍 From instruction-following to creative storytelling, dive into 2024's most impactful AI datasets! These gems are shaping everything from scientific research to video understanding.
