ML Research Hub
32.8K subscribers
4.35K photos
267 videos
23 files
4.7K links
Advancing research in Machine Learning – practical insights, tools, and techniques for researchers.

Admin: @HusseinSheikho || @Hussein_Sheikho
Download Telegram
πŸ€–πŸ§  Skyvern: The Future of Browser Automation Powered by AI and Computer Vision

πŸ—“οΈ 16 Nov 2025
πŸ“š AI News & Trends

In today’s fast-evolving digital landscape, automation plays a crucial role in enhancing productivity, efficiency and innovation. Yet, traditional browser automation tools often struggle with complexity, maintenance and reliability. They rely heavily on DOM parsing, XPaths and rigid scripts that easily break when websites change their layout. Enter Skyvern, an open-source, AI-driven browser automation platform developed ...

#Skyvern #BrowserAutomation #AIDriven #ComputerVision #OpenSource #WebAutomation
❀‍πŸ”₯1❀1πŸ‘1
✨P1: Mastering Physics Olympiads with Reinforcement Learning

πŸ“ Summary:
P1 is a family of open-source physics reasoning models trained via reinforcement learning. P1-235B-A22B achieved Gold-medal performance at IPhO 2025 and won 12 other competitions. These models also show strong generalizability on other reasoning tasks.

πŸ”Ή Publication Date: Published on Nov 17

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2511.13612
β€’ PDF: https://arxiv.org/pdf/2511.13612
β€’ Project Page: https://prime-rl.github.io/P1/
β€’ Github: https://github.com/PRIME-RL/P1

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#ReinforcementLearning #Physics #AI #MachineLearning #OpenSource
✨Instella: Fully Open Language Models with Stellar Performance

πŸ“ Summary:
Instella is a family of fully open language models trained on open data. It achieves state-of-the-art among fully open models and competes with leading open-weight LLMs. Specialized variants for long context and math reasoning are also offered.

πŸ”Ή Publication Date: Published on Nov 13

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2511.10628
β€’ PDF: https://arxiv.org/pdf/2511.10628
β€’ Github: https://github.com/AMD-AGI/Instella

πŸ”Ή Models citing this paper:
β€’ https://huggingface.co/amd/AMD-OLMo
β€’ https://huggingface.co/amd/Instella-3B-Instruct
β€’ https://huggingface.co/amd/Instella-3B

✨ Datasets citing this paper:
β€’ https://huggingface.co/datasets/amd/Instella-Long
β€’ https://huggingface.co/datasets/amd/Instella-GSM8K-synthetic

✨ Spaces citing this paper:
β€’ https://huggingface.co/spaces/DexterSptizu/AMD-OLMo-1B
β€’ https://huggingface.co/spaces/universeofml/DeepFocusTrain

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#LLMs #OpenSource #AI #MachineLearning #NLP
❀1
✨OpenUS: A Fully Open-Source Foundation Model for Ultrasound Image Analysis via Self-Adaptive Masked Contrastive Learning

πŸ“ Summary:
OpenUS is an open-source ultrasound foundation model built on a large public dataset. It uses a vision Mamba backbone and a novel self-adaptive masking framework to enhance pre-training, enabling label-efficient fine-tuning for various US tasks.

πŸ”Ή Publication Date: Published on Nov 14

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2511.11510
β€’ PDF: https://arxiv.org/pdf/2511.11510
β€’ Github: https://github.com/XZheng0427/OpenUS

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#OpenSource #FoundationModel #UltrasoundAI #MachineLearning #MedicalImaging
❀1
✨HunyuanVideo 1.5 Technical Report

πŸ“ Summary:
HunyuanVideo 1.5 is a lightweight, open-source video generation model achieving state-of-the-art visual quality and motion coherence. It employs an advanced DiT architecture with SSTA and an efficient video super-resolution network, enabling high-quality video creation on consumer GPUs.

πŸ”Ή Publication Date: Published on Nov 24

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2511.18870
β€’ PDF: https://arxiv.org/pdf/2511.18870
β€’ Github: https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#VideoGeneration #AI #DeepLearning #OpenSource #DiffusionModels
✨GigaEvo: An Open Source Optimization Framework Powered By LLMs And Evolution Algorithms

πŸ“ Summary:
GigaEvo is an open-source framework for LLM-guided evolutionary computation, providing modular tools for complex optimization. It enhances reproducibility of AlphaEvolve-inspired methods with detailed implementations, validated on challenging problems like Heilbronn triangle placement.

πŸ”Ή Publication Date: Published on Nov 17

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2511.17592
β€’ PDF: https://arxiv.org/pdf/2511.17592
β€’ Project Page: https://airi-institute.github.io/gigaevo-cover/
β€’ Github: https://github.com/FusionBrainLab/gigaevo-core

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#LLM #EvolutionaryAlgorithms #Optimization #OpenSource #AI
✨SWE-SQL: Illuminating LLM Pathways to Solve User SQL Issues in Real-World Applications

πŸ“ Summary:
SWE-SQL introduces BIRD-CRITIC, a new benchmark for SQL issue debugging, and Six-Gym, a training environment using f-Plan Boosting. Their open-source Bird-Fixer agent surpasses proprietary LLMs like GPT-4.1 in performance, democratizing advanced SQL-debugging capabilities.

πŸ”Ή Publication Date: Published on Jun 23

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2506.18951
β€’ PDF: https://arxiv.org/pdf/2506.18951
β€’ Project Page: https://bird-critic.github.io
β€’ Github: https://github.com/bird-bench/BIRD-CRITIC-1

✨ Datasets citing this paper:
β€’ https://huggingface.co/datasets/birdsql/bird-critic-1.0-flash-exp
β€’ https://huggingface.co/datasets/birdsql/bird-critic-1.0-open
β€’ https://huggingface.co/datasets/birdsql/bird-critic-1.0-postgresql

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#SQL #LLM #AI #Debugging #OpenSource
❀1
✨SWE-Bench++: A Framework for the Scalable Generation of Software Engineering Benchmarks from Open-Source Repositories

πŸ“ Summary:
SWE-Bench++ is an automated framework generating scalable, multilingual, repository-level coding tasks from live GitHub pull requests. It overcomes manual curation limits and static datasets, offering a benchmark to evaluate and improve code generation models across 11 languages.

πŸ”Ή Publication Date: Published on Dec 19

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2512.17419
β€’ PDF: https://arxiv.org/pdf/2512.17419
β€’ Project Page: https://research.turing.com/swebench
β€’ Github: https://huggingface.co/papers?q=GitHub%20pull%20requests

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#SoftwareEngineering #CodeGeneration #AIBenchmarking #MachineLearning #OpenSource
❀1
✨Simulstream: Open-Source Toolkit for Evaluation and Demonstration of Streaming Speech-to-Text Translation Systems

πŸ“ Summary:
Simulstream is an open-source toolkit for evaluating and demonstrating streaming speech-to-text translation. It supports long-form audio, incremental decoding, and re-translation, plus offers an interactive demo interface.

πŸ”Ή Publication Date: Published on Dec 19

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2512.17648
β€’ PDF: https://arxiv.org/pdf/2512.17648
β€’ Project Page: https://pypi.org/project/simulstream/

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#SpeechToText #MachineTranslation #NLP #OpenSource #StreamingAI
❀1
✨NVIDIA Nemotron 3: Efficient and Open Intelligence

πŸ“ Summary:
NVIDIA introduces Nemotron 3, a family of models with strong agentic, reasoning, and conversational capabilities. They feature a hybrid Mamba-Transformer MoE architecture for high throughput and long context, plus advanced post-training for tool use. The models will be openly released.

πŸ”Ή Publication Date: Published on Dec 24

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2512.20856
β€’ PDF: https://arxiv.org/pdf/2512.20856

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#AI #LLM #DeepLearning #NVIDIA #OpenSource
✨SciEvalKit: An Open-source Evaluation Toolkit for Scientific General Intelligence

πŸ“ Summary:
SciEvalKit is an open-source toolkit for evaluating AI models in science. It assesses scientific intelligence across diverse domains and competencies using expert-grade benchmarks and a flexible pipeline. This provides a standardized platform for scientific AI evaluation.

πŸ”Ή Publication Date: Published on Dec 26, 2025

πŸ”Ή Paper Links:
β€’ arXiv Page: https://arxiv.org/abs/2512.22334
β€’ PDF: https://arxiv.org/pdf/2512.22334

==================================

For more data science resources:
βœ“ https://t.me/DataScienceT

#AIevaluation #ScientificAI #OpenSource #AIBenchmarks #AIResearch