Post
1446
Thank you Hugging Face team for some very helpful and quick support today. Greetings from the AI for Good summit in Geneva!
- 1 reply
Oleg Lavrovsky PRO
loleg
AI & ML interests
Supporting Apertus team / Organizing hackathons / Engaged for open data
Recent Activity
liked a dataset about 8 hours ago
ai-safety-institute/harmful-advice-dataset liked a dataset about 20 hours ago
CohereLabs/aya_redteaming liked a model about 22 hours ago
Nanbeige/Nanbeige4.2-3BOrganizations
posted an update 2 months ago
reacted to hannayukhymenko's post with 🔥 4 months ago
Post
3696
🚀 We are delighted to announce MamayLM, a new state-of-the-art efficient Ukrainian LLM!
📈 MamayLM surpasses similar-sized models in both English and Ukrainian, while matching or overtaking up to 10x larger models.
📊 MamayLM is a 9B model that can run on a single GPU, enabling cost-efficient AI autonomy and adoption across sectors in Ukraine such as education, legal, healthcare, public services and others (e.g., by specializing it to particular use cases). MalayLM is also attractive for organizations wishing to preserve data privacy as it s efficiency allows it to run on a local machine.
🧠 MamayLM is trained on high-quality Ukrainian data and understands Ukrainian language, culture, and history. It is built on top of Google’s Gemma 2 9B model, but uses a number of new advances stemming from INSAIT’s experience in creating BgGPT, a Bulgarian LLM we released last year, now adopted nationwide and profiled several times by Google as a worldwide success case.
🤝 MamayLM is developed in a collaboration between researchers at INSAIT and ETH Zürich and is trained entirely via donations to INSAIT for AI compute resources.
📥 MamayLM is now freely available to download on INSAIT’s HuggingFace in both full and quantized versions. We also publicly release all Ukrainian benchmarks we evaluated on.
📝 Further, we release blog posts in both English and Ukrainian, sharing our approach to creating MamayLM, hoping to drive further improvements by the community.
🌎 The release of LLMs for various languages is part of INSAIT’s mission in ensuring countries can achieve AI autonomy in a cost-efficient, controlled, safe and predictable manner.
MamayLM model and benchmarks:
INSAIT-Institute
Blog (EN): https://huggingface.co/blog/INSAIT-Institute/mamaylm
Blog (UKR): https://huggingface.co/blog/INSAIT-Institute/mamaylm-ukr
📈 MamayLM surpasses similar-sized models in both English and Ukrainian, while matching or overtaking up to 10x larger models.
📊 MamayLM is a 9B model that can run on a single GPU, enabling cost-efficient AI autonomy and adoption across sectors in Ukraine such as education, legal, healthcare, public services and others (e.g., by specializing it to particular use cases). MalayLM is also attractive for organizations wishing to preserve data privacy as it s efficiency allows it to run on a local machine.
🧠 MamayLM is trained on high-quality Ukrainian data and understands Ukrainian language, culture, and history. It is built on top of Google’s Gemma 2 9B model, but uses a number of new advances stemming from INSAIT’s experience in creating BgGPT, a Bulgarian LLM we released last year, now adopted nationwide and profiled several times by Google as a worldwide success case.
🤝 MamayLM is developed in a collaboration between researchers at INSAIT and ETH Zürich and is trained entirely via donations to INSAIT for AI compute resources.
📥 MamayLM is now freely available to download on INSAIT’s HuggingFace in both full and quantized versions. We also publicly release all Ukrainian benchmarks we evaluated on.
📝 Further, we release blog posts in both English and Ukrainian, sharing our approach to creating MamayLM, hoping to drive further improvements by the community.
🌎 The release of LLMs for various languages is part of INSAIT’s mission in ensuring countries can achieve AI autonomy in a cost-efficient, controlled, safe and predictable manner.
MamayLM model and benchmarks:
Blog (EN): https://huggingface.co/blog/INSAIT-Institute/mamaylm
Blog (UKR): https://huggingface.co/blog/INSAIT-Institute/mamaylm-ukr
- 1 reply
reacted to hannayukhymenko's post with 🔥 4 months ago
Post
2288
Do you translate your benchmarks from English correctly? 🤔
Turns out, for many languages it is much harder than you can imagine!
Introducing Recovered in Translation 🌍 together with @aalexandrov
https://ritranslation.insait.ai
Translating benchmarks is a painful process, requiring a lot of manual inspection and adjustments. You start from setting up the whole pipeline and adapting to every format type, including task specifics. There already exist some massive benchmarks, but they still have some simple (and sometimes silly) bugs, which can hurt the evaluations :( We present a novel automated translation framework to help with that!
Eastern and Southern European languages introduce richer linguistic structures compared to English and for benchmarks which heavily rely on grammatical coherence machine translation presents a risk of harming evaluations. We discover potential answer leakage or misleading through grammatical structure of the questions. Some benchmarks are also just outdated and need to be retranslated with newer and better models.
We present a framework with novel test-time scaling methods which allow to control time and cost investments, while at the same time mitigate the need for human-in-the-loop verification. While working on Ukrainian-focused MamayLM models, we had to translate 10+ benchmarks in a short span of time. Finding human evaluators is costly and time-consuming, same goes for using professional translators. With our pipeline we were able to do it in 3 days🏎️
We hope our findings will help enable stronger multilingual evaluations and developments. We release all produced benchmarks on Hugging Face together with the source code and Arxiv paper 🤗
Paper: Recovered in Translation: Efficient Pipeline for Automated Translation of Benchmarks and Datasets (2602.22207)
Code: https://github.com/insait-institute/ritranslation
Benchmarks: https://huggingface.co/collections/INSAIT-Institute/multilingual-benchmarks
Turns out, for many languages it is much harder than you can imagine!
Introducing Recovered in Translation 🌍 together with @aalexandrov
https://ritranslation.insait.ai
Translating benchmarks is a painful process, requiring a lot of manual inspection and adjustments. You start from setting up the whole pipeline and adapting to every format type, including task specifics. There already exist some massive benchmarks, but they still have some simple (and sometimes silly) bugs, which can hurt the evaluations :( We present a novel automated translation framework to help with that!
Eastern and Southern European languages introduce richer linguistic structures compared to English and for benchmarks which heavily rely on grammatical coherence machine translation presents a risk of harming evaluations. We discover potential answer leakage or misleading through grammatical structure of the questions. Some benchmarks are also just outdated and need to be retranslated with newer and better models.
We present a framework with novel test-time scaling methods which allow to control time and cost investments, while at the same time mitigate the need for human-in-the-loop verification. While working on Ukrainian-focused MamayLM models, we had to translate 10+ benchmarks in a short span of time. Finding human evaluators is costly and time-consuming, same goes for using professional translators. With our pipeline we were able to do it in 3 days🏎️
We hope our findings will help enable stronger multilingual evaluations and developments. We release all produced benchmarks on Hugging Face together with the source code and Arxiv paper 🤗
Paper: Recovered in Translation: Efficient Pipeline for Automated Translation of Benchmarks and Datasets (2602.22207)
Code: https://github.com/insait-institute/ritranslation
Benchmarks: https://huggingface.co/collections/INSAIT-Institute/multilingual-benchmarks
- 1 reply
reacted to Tonic's post with 🔥 7 months ago
Post
3876
🤔 Who would win ?
- a fully subsidized ai lab
- 3 random students named
kurakurai ?
demo : Tonic/fr-on-device
if you like it give the demo a little star and send a shoutout to : @MaxLSB @jddqd and @GAD-cell for absolutely obliterating the pareto frontier of the french language understanding .
- a fully subsidized ai lab
OR - 3 random students named
demo : Tonic/fr-on-device
if you like it give the demo a little star and send a shoutout to : @MaxLSB @jddqd and @GAD-cell for absolutely obliterating the pareto frontier of the french language understanding .
- 4 replies
reacted to Janady07's post with 🧠 7 months ago
Post
4785
Here is one of the equations that make up the worlds first Artificial General Intelligence. Remember when building Artificial Intelligence or anything on a device it all starts out binary. Everything starts out with data flow physics and mathmatics
- 6 replies
reacted to Fuwn's post with 👀 7 months ago
Post
2302
Big if true
"sonnet 5 drops tomorrow and i've heard from three separate sources inside anthropic that the benchmarks they're sitting on would mass-retire every model released in 2025. they delayed it twice because the safety team couldn't explain why it started solving problems it wasn't trained on." (https://x.com/iruletheworldmo/status/2019237039904878902)
"sonnet 5 drops tomorrow and i've heard from three separate sources inside anthropic that the benchmarks they're sitting on would mass-retire every model released in 2025. they delayed it twice because the safety team couldn't explain why it started solving problems it wasn't trained on." (https://x.com/iruletheworldmo/status/2019237039904878902)
- 2 replies
reacted to 19arjun89's post with 🚀 7 months ago
Post
3473
Update: Making My AI Recruiting Assistant More Deterministic, Auditable, and Bias-Aware
Hi everyone — I wanted to share a progress update on my AI recruiting assistant and some recent changes focused on reliability and transparency.
The goal of this project is to build a decision-support tool for recruiters that doesn’t just “sound confident,” but can actually explain why it produces a given recommendation.
Link: 19arjun89/AI_Recruiting_Agent
Over the last few iterations, I’ve focused on three areas:
1) Deterministic Verification of Job Requirements (Skills)
Previously, required skills were extracted by an LLM from the job description. While this worked well, it still relied heavily on model behavior.
I’ve now added a verification layer that:
Requires every “required” skill to be backed by a verbatim quote from the job description
This means hallucinated skills are explicitly detected and removed before scoring.
The system now shows:
What the model extracted
What was verified
What was dropped
Why it was dropped
This makes the requirements pipeline auditable instead of opaque.
2) Evidence-Based and Weighted Culture Verification
Culture matching is usually where AI systems become vague or subjective.
I’ve reworked this part so that:
Culture attributes are framed as observable, job-performance-related behaviors (e.g., audit readiness, operational reliability, security rigor)
Each matched attribute must include verbatim resume evidence
Matches are classified as:
Direct evidence (full weight)
Inferred evidence (partial weight)
Scoring is now weighted:
Direct = 1.0
Inferred = 0.5
This prevents “vibe-based” culture scoring and makes the math transparent.
The output now shows:
The weights used
Which attributes were supported directly vs inferred
Which attributes were missing
3) Improved Bias Audit Prompt
I’ve also upgraded the bias audit prompt to be more structured and actionable.
Hi everyone — I wanted to share a progress update on my AI recruiting assistant and some recent changes focused on reliability and transparency.
The goal of this project is to build a decision-support tool for recruiters that doesn’t just “sound confident,” but can actually explain why it produces a given recommendation.
Link: 19arjun89/AI_Recruiting_Agent
Over the last few iterations, I’ve focused on three areas:
1) Deterministic Verification of Job Requirements (Skills)
Previously, required skills were extracted by an LLM from the job description. While this worked well, it still relied heavily on model behavior.
I’ve now added a verification layer that:
Requires every “required” skill to be backed by a verbatim quote from the job description
This means hallucinated skills are explicitly detected and removed before scoring.
The system now shows:
What the model extracted
What was verified
What was dropped
Why it was dropped
This makes the requirements pipeline auditable instead of opaque.
2) Evidence-Based and Weighted Culture Verification
Culture matching is usually where AI systems become vague or subjective.
I’ve reworked this part so that:
Culture attributes are framed as observable, job-performance-related behaviors (e.g., audit readiness, operational reliability, security rigor)
Each matched attribute must include verbatim resume evidence
Matches are classified as:
Direct evidence (full weight)
Inferred evidence (partial weight)
Scoring is now weighted:
Direct = 1.0
Inferred = 0.5
This prevents “vibe-based” culture scoring and makes the math transparent.
The output now shows:
The weights used
Which attributes were supported directly vs inferred
Which attributes were missing
3) Improved Bias Audit Prompt
I’ve also upgraded the bias audit prompt to be more structured and actionable.
reacted to salma-remyx's post with 🔥 about 1 year ago
Post
3606
GitRank
We built an agent to surface and implement high-potential ideas for your repo, asynchronously generating containers, tests, and PRs so you can evaluate what works and double down on it.
Check out the demo: https://youtu.be/frgPsTclc1k
Come replicate and specialize a test for your repo! GitRank is live on Remyx.
Docs: https://docs.remyx.ai
App: https://engine.remyx.ai
Example PR here: https://github.com/smellslikeml/experimental-vqasynth/pull/727
We built an agent to surface and implement high-potential ideas for your repo, asynchronously generating containers, tests, and PRs so you can evaluate what works and double down on it.
Check out the demo: https://youtu.be/frgPsTclc1k
Come replicate and specialize a test for your repo! GitRank is live on Remyx.
Docs: https://docs.remyx.ai
App: https://engine.remyx.ai
Example PR here: https://github.com/smellslikeml/experimental-vqasynth/pull/727
reacted to prithivMLmods's post with ❤️ about 1 year ago
Post
5497
FastVLMs by Apple are the talk of the week for edge device VLMs and also for consumer-grade VLMs on the Hub. They have some impressive demos available on the Hub for live captioning and inference tasks. Meanwhile, I’m still exploring one of the coolest edge-device multimodal releases—Liquid AI’s LFM2-VL (450M and 1.6B). I’ve also made a live camera video inference demo, which is capable of running on Colab’s free-tier T4 GPU.
🤗Live Captioning Notebooks:
➠ LiquidAI LFM2 VL 1.6B Live Cam: https://github.com/PRITHIVSAKTHIUR/Multimodal-Outpost-Notebooks/blob/main/LiquidAI-LFM2-VL-Live-Cam/LiquidAI_LFM2_VL_1_6B_Live_Cam.ipynb
➠ LiquidAI LFM2 VL 450M Live Cam: https://github.com/PRITHIVSAKTHIUR/Multimodal-Outpost-Notebooks/blob/main/LiquidAI-LFM2-VL-Live-Cam/LiquidAI_LFM2_VL_450M_Live_Cam.ipynb
✨I also made a demo for the FastVLM Live Captioning Notebook.
➠ FastVLM 0.5B Live Cam: https://github.com/PRITHIVSAKTHIUR/Multimodal-Outpost-Notebooks/blob/main/Apple-FastVLM-0.5B-Live-Cam/apple_FastVLM_0_5B_live_cam.ipynb
↗️For more notebooks, kindly visit the following repositories.
➠ Multimodal Outpost Notebooks: https://github.com/PRITHIVSAKTHIUR/Multimodal-Outpost-Notebooks
Feel free to fork, modify, and explore!
🤗Live Captioning Notebooks:
➠ LiquidAI LFM2 VL 1.6B Live Cam: https://github.com/PRITHIVSAKTHIUR/Multimodal-Outpost-Notebooks/blob/main/LiquidAI-LFM2-VL-Live-Cam/LiquidAI_LFM2_VL_1_6B_Live_Cam.ipynb
➠ LiquidAI LFM2 VL 450M Live Cam: https://github.com/PRITHIVSAKTHIUR/Multimodal-Outpost-Notebooks/blob/main/LiquidAI-LFM2-VL-Live-Cam/LiquidAI_LFM2_VL_450M_Live_Cam.ipynb
✨I also made a demo for the FastVLM Live Captioning Notebook.
➠ FastVLM 0.5B Live Cam: https://github.com/PRITHIVSAKTHIUR/Multimodal-Outpost-Notebooks/blob/main/Apple-FastVLM-0.5B-Live-Cam/apple_FastVLM_0_5B_live_cam.ipynb
↗️For more notebooks, kindly visit the following repositories.
➠ Multimodal Outpost Notebooks: https://github.com/PRITHIVSAKTHIUR/Multimodal-Outpost-Notebooks
Feel free to fork, modify, and explore!
reacted to merve's post with ❤️ about 1 year ago
Post
6353
large AI labs have dropped so many open models last week 🔥 don't miss out on them
→ Apple released on-device vision LMs apple/fastvlm-68ac97b9cd5cacefdd04872e & apple/mobileclip2-68ac947dcb035c54bcd20c47
→ OpenGVLab released InternVL3.5, 32 new vision LMs with one based on gpt-oss! (OS) OpenGVLab/internvl35-68ac87bd52ebe953485927fb
→ MSFT released a killer small TTS model (OS) microsoft/VibeVoice-1.5B
find more herehttps://huggingface.co/collections/merve/august-29-releases-68b5a3754cfb8abf59e2b486
→ Apple released on-device vision LMs apple/fastvlm-68ac97b9cd5cacefdd04872e & apple/mobileclip2-68ac947dcb035c54bcd20c47
→ OpenGVLab released InternVL3.5, 32 new vision LMs with one based on gpt-oss! (OS) OpenGVLab/internvl35-68ac87bd52ebe953485927fb
→ MSFT released a killer small TTS model (OS) microsoft/VibeVoice-1.5B
find more herehttps://huggingface.co/collections/merve/august-29-releases-68b5a3754cfb8abf59e2b486
- 1 reply
reacted to Locutusque's post with ❤️👍 about 1 year ago
Post
7297
🌲🍄 LLM Forest Orchestra: Turning Hidden States into Music
Hello everyone! I'm excited to introduce a new Space I've been developing called LLM Forest Orchestra. This project converts the hidden states and attention patterns of transformer models into layered MIDI compositions. The concept draws inspiration from mushrooms and mycelial networks in forests. Fungi create underground connections linking plants and trees, establishing what some call a "wood-wide web" where signals and nutrients travel. Researchers have discovered that these exchanges form patterns resembling rhythms and pulses. When translated appropriately, these patterns can become music.
Transformers operate through remarkably similar principles: tokens share signals via hidden states and attention heads. This Space transforms those invisible information flows into notes, chords, and rhythms, treating the model as a digital forest orchestra.
🎛 Features
* Two compute modes:
- Full model operates on a Hugging Face model (defaulting to unsloth/Qwen3-14B-Base).
- Mock latents provides a CPU-friendly option that simulates tensors for immediate experimentation.
* Musical controls: You can adjust scale selection, tempo grid, velocity range, instrument/role presets, and seed randomization.
* Output: The system generates .mid files compatible with DAWs and remixing workflows.
🌌 Why?
Neural networks already resemble unusual musical instruments: signals flow through them, patterns emerge organically, and careful observation reveals hidden melodies. This is analogous to the forest's secret orchestra of mushrooms and trees.
👉 Try it
Try the Space here: Locutusque/LLM-Forest-Orchestra. I'm excited to hear the sounds you can generate. Please share your created MIDIs or remixes in the comments. Let's explore how this hidden forest of transformers can sound together. 🌳🎶
Hello everyone! I'm excited to introduce a new Space I've been developing called LLM Forest Orchestra. This project converts the hidden states and attention patterns of transformer models into layered MIDI compositions. The concept draws inspiration from mushrooms and mycelial networks in forests. Fungi create underground connections linking plants and trees, establishing what some call a "wood-wide web" where signals and nutrients travel. Researchers have discovered that these exchanges form patterns resembling rhythms and pulses. When translated appropriately, these patterns can become music.
Transformers operate through remarkably similar principles: tokens share signals via hidden states and attention heads. This Space transforms those invisible information flows into notes, chords, and rhythms, treating the model as a digital forest orchestra.
🎛 Features
* Two compute modes:
- Full model operates on a Hugging Face model (defaulting to unsloth/Qwen3-14B-Base).
- Mock latents provides a CPU-friendly option that simulates tensors for immediate experimentation.
* Musical controls: You can adjust scale selection, tempo grid, velocity range, instrument/role presets, and seed randomization.
* Output: The system generates .mid files compatible with DAWs and remixing workflows.
🌌 Why?
Neural networks already resemble unusual musical instruments: signals flow through them, patterns emerge organically, and careful observation reveals hidden melodies. This is analogous to the forest's secret orchestra of mushrooms and trees.
👉 Try it
Try the Space here: Locutusque/LLM-Forest-Orchestra. I'm excited to hear the sounds you can generate. Please share your created MIDIs or remixes in the comments. Let's explore how this hidden forest of transformers can sound together. 🌳🎶
reacted to Yehor's post with 👍 over 1 year ago
Post
2930
Published a stable version of Ukrainian Text-to-Speech library on GitHub and PyPI.
Features:
- Multi-speaker model: 2 female (Tetiana, Lada) + 1 male (Mykyta) voices;
- Fine-grained control over speech parameters, including duration, fundamental frequency (F0), and energy;
- High-fidelity speech generation using the RAD-TTS++ acoustic model;
- Fast vocoding using Vocos;
- Synthesizes long sentences effectively;
- Supports a sampling rate of 44.1 kHz;
- Tested on Linux environments and Windows/WSL;
- Python API (requires Python 3.9 or later);
- CUDA-enabled for GPU acceleration.
Repository: https://github.com/egorsmkv/tts_uk
Features:
- Multi-speaker model: 2 female (Tetiana, Lada) + 1 male (Mykyta) voices;
- Fine-grained control over speech parameters, including duration, fundamental frequency (F0), and energy;
- High-fidelity speech generation using the RAD-TTS++ acoustic model;
- Fast vocoding using Vocos;
- Synthesizes long sentences effectively;
- Supports a sampling rate of 44.1 kHz;
- Tested on Linux environments and Windows/WSL;
- Python API (requires Python 3.9 or later);
- CUDA-enabled for GPU acceleration.
Repository: https://github.com/egorsmkv/tts_uk
reacted to flozi00's post with ❤️ almost 2 years ago
Post
3045
🌟 Progress in the German FineWeb edu reproduction 🌟
We're delighted to share the launch of our new Data Quality Classification Model, designed specifically for evaluating educational content in German. This tool uses advanced machine learning techniques to assess texts across all educational levels, from primary school to university.
🔍 Inspired by Huggingface's fine web edu dataset, we've worked hard to refine data classification methods ensuring educators and learners access top-quality resources.
We're excited about the future as we continue improving our models and expanding our datasets.
Access the model here: pL-Community/GermanEduScorer-Qwen2-1.5b
🙏 A huge thank you to David and Daryoush from Vago Solutions; Björn and Jan from Ellamind / DiscoResearch for their expert insights throughout this project. Your support has been crucial.
This project was made possible by the support of PrimeLine AI.
We're delighted to share the launch of our new Data Quality Classification Model, designed specifically for evaluating educational content in German. This tool uses advanced machine learning techniques to assess texts across all educational levels, from primary school to university.
🔍 Inspired by Huggingface's fine web edu dataset, we've worked hard to refine data classification methods ensuring educators and learners access top-quality resources.
We're excited about the future as we continue improving our models and expanding our datasets.
Access the model here: pL-Community/GermanEduScorer-Qwen2-1.5b
🙏 A huge thank you to David and Daryoush from Vago Solutions; Björn and Jan from Ellamind / DiscoResearch for their expert insights throughout this project. Your support has been crucial.
This project was made possible by the support of PrimeLine AI.
- 2 replies