Artificial Intelligence l l AI Updates
Kanalga Telegramโda oโtish
News about AI & DL & ML!!! Admin: @Gayrat_Tangriberganov
Ko'proq ko'rsatish1 615
Obunachilar
+124 soatlar
+27 kunlar
+430 kunlar
Postlar arxiv
๐ GitHub_Link
โ๏ธ ML-powered speech recognition directly in your browser
#STT
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ A react frontend for segmentation and subject extraction using Cartoon Segmentation ๐ฅ
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ FastVLM: Efficient Vision Encoding for Vision Language Models
#VLMs
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ TransPixeler: Advancing Text-to-Video Generation with Transparency ๐ฅ
#txt2vid #img2vid
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ๐ฅ Fine-tuning Qwen2.5-VL-3B๐ฅ
#LLMs
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ HiFi-Score: Fine-grained Image Description Evaluation with Hierarchical Parsing Graphs ๐ฅ
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ Count Anything ๐ฅ๐ฅ๐ฅ
#SAM
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ CFG-Zero*: Improved Classifier-Free Guidance for Flow Matching Models ๐ฅ
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ MLflow for Machine Learning Development
#Tutorial
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ CoMotion: Concurrent Multi-person 3D Motion
#3D
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ Event-Enhanced Blurry Video Super-Resolution ๐ฅ๐ฅ๐ฅ
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ AirCanvas: it is is a computer vision project that lets you draw using hand gestures.
#Tutorial
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ Sources
Advances and Challenges in Foundation Agents
https://arxiv.org/pdf/2504.01990v1
Why do Multi-agent systems fail
https://arxiv.org/pdf/2503.13657
API vs. GUI Agents - Divergence and Convergence
https://arxiv.org/html/2503.11069v1
PaperBench - Evaluating AI's Ability to Replicate Research
https://arxiv.org/abs/2504.01848
MemInsight: Autonomous Memory Augmentation for LLM Agents
https://arxiv.org/pdf/2503.21760
BEARCUBS: A Benchmark for Computer-Using Web Agents
https://arxiv.org/pdf/2503.07919
AgentRxiv: Towards Collaborative Autonomous Research
https://arxiv.org/abs/2503.18102
PLAY2PROMPT: Zero-shot Tool Instruction Optimization for LLM Agents
https://arxiv.org/abs/2503.14432
Agents Play Thousands of 3D Video Games using PORTAL
https://arxiv.org/abs/2503.13356
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ DSO: Aligning 3D Generators with Simulation Feedback for Physical Soundness
#3D
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstruction
#4D
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ Packing Input Frame Context in Next-Frame Prediction Models for Video Generation ๐ฅ๐ฅ๐ฅ
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ Pusa: Thousands Timesteps Video Diffusion Model
#txt2img #img2vid
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ ๐ฅ Spio ๐ฅ: Memory-Efficient CUDA kernels for training ConvNets with PyTorch
#Cuda
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
๐ฅ Kornia YOLO in Rust ๐ฅ
๐ ๐ High-performance object detection with YOLOv8 using Kornia and Candle
๐ ๐ง Multiple model sizes (n, s, m, l, x) for different performance/accuracy tradeoffs
๐ ๐ฆ Pre-trained models automatically downloaded from Hugging Face
๐ ๐ผ Simple API for inference on images
๐ ๐งฎ Non-maximum suppression for refined detection results
๐ ๐ Built on the Kornia Rust computer vision ecosystem
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
๐ GitHub_Link
โ๏ธ AutoVFX: Physically Realistic Video Editing from Natural Language Instructions.
Join my channel:
๐๐๐๐๐๐
https://t.me/Artificial_Intelligence_Updates
Endi mavjud! Telegram Tadqiqoti 2025 โ yilning asosiy insaytlari 
