ML Research Hub – Telegram
ML Research Hub
32.7K subscribers
4.03K photos
230 videos
23 files
4.34K links
Advancing research in Machine Learning – practical insights, tools, and techniques for researchers.

Admin: @HusseinSheikho || @Hussein_Sheikho
Download Telegram
How to Evaluate Speech Translation with Source-Aware Neural MT Metrics

📝 Summary:
This study introduces source-aware metrics for speech translation evaluation by generating text proxies from audio, like ASR trannoscripts or back-translations. A new re-segmentation algorithm resolves alignment issues. These methods improve evaluation accuracy for speech translation systems.

🔹 Publication Date: Published on Nov 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2511.03295
• PDF: https://arxiv.org/pdf/2511.03295

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT

#SpeechTranslation #NMTMetrics #ASR #NLP #DeepLearning
RDMA Point-to-Point Communication for LLM Systems

📝 Summary:
TransferEngine provides a uniform interface for flexible point-to-point communication in LLM systems, overcoming NIC-specific limitations. It bridges different hardware, providing high throughput for disaggregated inference, RL, and MoE tasks. This solution avoids hardware lock-in and complements...

🔹 Publication Date: Published on Oct 31

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.27656
• PDF: https://arxiv.org/pdf/2510.27656
• Github: https://github.com/perplexityai/pplx-garden

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT

#RDMA #LLM #HPC #AIInfrastructure #DistributedSystems
SAIL-RL: Guiding MLLMs in When and How to Think via Dual-Reward RL Tuning

📝 Summary:
SAIL-RL uses a dual-reward RL system to teach MLLMs when and how to think. This improves reasoning, reduces hallucinations, and achieves competitive performance against commercial models like GPT-4o.

🔹 Publication Date: Published on Nov 4

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2511.02280
• PDF: https://arxiv.org/pdf/2511.02280

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT

#MLLMs #ReinforcementLearning #AI #GenerativeAI #DeepLearning
SIMS-V: Simulated Instruction-Tuning for Spatial Video Understanding

📝 Summary:
SIMS-V uses 3D simulators to generate diverse spatial video training data. This efficiently trains multimodal language models, overcoming real-world data bottlenecks. A 7B model trained on this simulated data significantly outperforms larger baselines on real-world spatial reasoning tasks.

🔹 Publication Date: Published on Nov 6

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2511.04668
• PDF: https://arxiv.org/pdf/2511.04668
• Github: https://ellisbrown.github.io/sims-v/

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT

#SpatialAI #MultimodalLLM #SimulatedData #ComputerVision #DeepLearning
EVTAR: End-to-End Try on with Additional Unpaired Visual Reference

📝 Summary:
EVTAR is an end-to-end virtual try-on model that enhances accuracy and garment detail preservation using additional reference images. It simplifies the process by requiring only source and target garment inputs, producing high-quality, realistic try-on results.

🔹 Publication Date: Published on Nov 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2511.00956
• PDF: https://arxiv.org/pdf/2511.00956
• Github: https://github.com/360CVGroup/EVTAR

🔹 Models citing this paper:
https://huggingface.co/qihoo360/EVTAR

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT

#VirtualTryOn #ComputerVision #DeepLearning #AIFashion #ImageSynthesis
EVODiff: Entropy-aware Variance Optimized Diffusion Inference

📝 Summary:
EVODiff optimizes diffusion model inference using an entropy-aware variance method. It leverages information theory to reduce uncertainty and minimize errors. This approach significantly outperforms gradient-based solvers, enhancing efficiency and reconstruction quality.

🔹 Publication Date: Published on Sep 30

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.26096
• PDF: https://arxiv.org/pdf/2509.26096
• Project Page: https://neurips.cc/virtual/2025/poster/115792
• Github: https://github.com/ShiguiLi/EVODiff

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT

#DiffusionModels #DeepLearning #MachineLearning #Optimization #InformationTheory
1
🤖🧠 DeepSeek-V3: Pioneering Large-Scale AI Efficiency and Open Innovation

🗓️ 07 Nov 2025
📚 AI News & Trends

The field of artificial intelligence has entered a transformative phase – one defined by scale, specialization and accessibility. As the demand for larger and more capable language models grows, the challenge lies not only in achieving state-of-the-art performance but also in doing so efficiently and sustainably. DeepSeek-AI’s latest release, DeepSeek-V3 redefines what is possible at ...

#DeepSeekV3 #AIInnovation #LargeScaleAI #OpenInnovation #ArtificialIntelligence #AIEfficiency
🤖🧠 olmOCR: Redefining Document Understanding with Vision-Language Models

🗓️ 07 Nov 2025
📚 AI News & Trends

The digital era has seen an explosion in the amount of information stored in PDFs, scanned documents and image-based files. From research papers and corporate reports to handwritten notes and invoices, these unstructured sources hold trillions of valuable data points. Yet, extracting and converting this data into structured, machine-readable text has long been a challenge. ...

#olmOCR #DocumentUnderstanding #VisionLanguageModels #AIInnovation #UnstructuredData #DigitalTransformation
🤖🧠 FIBO: The First JSON-Native, Open-Source Text-to-Image Model Built for Real-World Control and Accuracy

🗓️ 07 Nov 2025
📚 AI News & Trends

The world of generative AI has evolved rapidly with text-to-image tools enabling creators, marketers, designers and enterprises to bring ideas to life with unprecedented ease. However, most existing models have a clear limitation: they prioritize imagination at the cost of control. Whether producing inconsistent styles, unpredictable lighting or drifting away from user prompts, traditional models ...

#FIBO #TextToImage #GenerativeAI #OpenSource #JSONNative #RealWorldControl
OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM

📝 Summary:
OmniVinci is an open-source omni-modal LLM that improves cross-modal understanding for audio, vision, and robotics. It features innovative architecture for better embedding alignment and temporal capture, along with efficient data curation. OmniVinci outperforms competitors while using significan...

🔹 Publication Date: Published on Oct 17

🔹 Paper Links:
• arXiv Page: https://arxivexplained.com/papers/omnivinci-enhancing-architecture-and-data-for-omni-modal-understanding-llm
• PDF: https://arxiv.org/pdf/2510.15870
• Project Page: https://nvlabs.github.io/OmniVinci/
• Github: https://github.com/NVlabs/OmniVinci

🔹 Models citing this paper:
https://huggingface.co/nvidia/omnivinci

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT

#LLM #MultimodalAI #Robotics #DeepLearning #OpenSource
EdgeTAM: On-Device Track Anything Model

📝 Summary:
EdgeTAM optimizes SAM 2 for mobile devices by addressing memory attention bottlenecks with a novel 2D Spatial Perceiver. This lightweight Transformer encodes frame-level memories to reduce computational cost. A distillation pipeline improves performance, enabling high-quality video segmentation a...

🔹 Publication Date: Published on Jan 13

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2501.07256
• PDF: https://arxiv.org/pdf/2501.07256
• Github: https://github.com/facebookresearch/edgetam

🔹 Models citing this paper:
https://huggingface.co/yonigozlan/EdgeTAM-hf
https://huggingface.co/facebook/EdgeTAM

Spaces citing this paper:
https://huggingface.co/spaces/merve/EdgeTAM
https://huggingface.co/spaces/yonigozlan/edgetam
https://huggingface.co/spaces/facebook/EdgeTAM

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT

#EdgeAI #VideoSegmentation #ComputerVision #MobileAI #DeepLearning
1
🤖🧠 Kimi Linear: The Future of Efficient Attention in Large Language Models

🗓️ 08 Nov 2025
📚 AI News & Trends

The rapid evolution of large language models (LLMs) has unlocked new capabilities in natural language understanding, reasoning, coding and multimodal tasks. However, as models grow more advanced, one major challenge persists: computational efficiency. Traditional full-attention architectures struggle to scale efficiently, especially when handling long context windows and real-time inference workloads. The increasing demand for agent-like ...

#KimiLinear #EfficientAttention #LargeLanguageModels #LLM #ComputationalEfficiency #AIInnovation
🤖🧠 Meilisearch: The Lightning-Fast, AI-Ready Search Engine for Modern Applications

🗓️ 08 Nov 2025
📚 AI News & Trends

Search is no longer a luxury feature. Today’s users expect instant, relevant results across e-commerce platforms, SaaS tools, media libraries and knowledge systems. With AI-powered experiences becoming the new standard, developers need search infrastructure that is fast, flexible, developer-friendly and ready for hybrid semantic search. This is where Meilisearch stands out. Meilisearch is an open-source, ...

#Meilisearch #AIReadySearch #LightningFast #SearchEngine #ModernApplications #OpenSource
🤖🧠 Pixeltable: The Future of Declarative Data Infrastructure for Multimodal AI Workloads

🗓️ 08 Nov 2025
📚 AI News & Trends

In the rapidly evolving AI landscape, building intelligent applications is no longer just about having powerful models. The real challenge lies in handling complex data pipelines, integrating multiple systems and scaling multimodal workloads efficiently. Traditional AI app development stacks involve databases, vector stores, ETL pipelines, model serving layers, orchestration tools, caching systems and lineage tracking ...

#Pixeltable #DeclarativeDataInfrastructure #MultimodalAI #AIDevelopment #DataPipelines #AIWorkloads
🤖🧠 Chandra OCR: The Future of Document Understanding and Layout-Aware Text Extraction

🗓️ 08 Nov 2025
📚 AI News & Trends

Optical Character Recognition (OCR) has evolved far beyond simply converting scanned text into digital characters. With the rise of artificial intelligence and large language models, the industry is shifting toward intelligent document understanding where structure, context and visual elements matter as much as the text itself. In this landscape, Chandra emerges as a breakthrough solution. ...

#ChandraOCR #DocumentUnderstanding #LayoutAwareText #OpticalCharacterRecognition #AIDocumentProcessing #IntelligentOCR
🤖🧠 LMCache: Accelerating LLM Inference With Next-Generation KV Cache Technology

🗓️ 08 Nov 2025
📚 AI News & Trends

As large language models (LLMs) continue to scale in size and complexity, organizations face an increasingly critical challenge: serving models efficiently in real-world applications. While LLM capabilities are rapidly evolving, the bottleneck of inference performance remains a major limitation especially when dealing with long-context workloads or high-traffic enterprise environments. This is where LMCache steps in. ...

#LMCache #LLMInference #KVCache #LargeLanguageModels #AIAcceleration #NextGenTechnology
🤖🧠 Dify: A Powerful #1 Production-Ready Platform for Building Advanced LLM Applications

🗓️ 08 Nov 2025
📚 AI News & Trends

The rapid growth of AI has made large language models (LLMs) an essential component for automation, content creation, data intelligence and workflow optimization. But moving AI concepts from prototype to production has traditionally required significant engineering effort, infrastructure planning and model-orchestration expertise. Dify changes that entirely. Dify is an open-source platform designed to help developers, ...

#Dify #LLMApplications #ProductionReady #AIPower #LargeLanguageModels #OpenSourcePlatform
JoyAgent-JDGenie: Technical Report on the GAIA

📝 Summary:
This paper introduces JoyAgent-JDGenie, a generalist AI agent architecture. It integrates multi-agent planning, hierarchical memory, and advanced tools to achieve superior performance across diverse tasks, outperforming baselines and approaching proprietary systems.

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.00510
• PDF: https://arxiv.org/pdf/2510.00510
• Github: https://github.com/jd-opensource/joyagent-jdgenie

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT

#AIAgent #GeneralistAI #MultiAgentSystems #AIResearch #MachineLearning
nature papers: 2000$

Q1 and  Q2 papers    1000$

Q3 and Q4 papers   500$

Doctoral thesis (complete)    700$

M.S thesis         300$

paper simulation   200$

Contact me @husseinsheikho
🤖🧠 Open WebUI: The Most Powerful Self-Hosted AI Platform for Local and Private LLMs

🗓️ 09 Nov 2025
📚 AI News & Trends

In the rapidly evolving landscape of artificial intelligence, the ability to run large language models securely and efficiently has become a major priority for developers, enterprises and privacy-focused users. While cloud-based AI services are convenient, they rely heavily on remote servers, internet access and third-party control. This is where Open WebUI stands out as a ...

#OpenWebUI #SelfHostedAI #PrivateLLMs #LocalAI #AISecurity #OpenSourcePlatform