#deeplearning

@dev_community Section 4a: From the variational bound to a trainable min-max game #MachineLearning #DeepLearning #AI

@dev_community Vertical Models Just Stole the Inference Stack #AI #MachineLearning #DeepLearning

@habr_ru Nano Banana 2.1 или Nano Banana 2: что лучше? Сравниваем нейросети для генерации изображений на пяти одинаковых промптах #AI #ImageGeneration #DeepLearning

@dev_community An analysis of 5 key papers #AI #MachineLearning #DeepLearning

@dev_community You could have invented transformers from memory #AI #MachineLearning #DeepLearning

@aws_builder_center Transformers in Myanmar - Advancing Burmese Language AI Through Modern Deep Learning #AI #DeepLearning #Cloud

@bloomberg_technology DeepSeek Pushes Out Updated AI Model It Says Is Faster, Leaner #AI #Tech #DeepLearning

@dev_community DeepSeek V4.1 Flash Just Made Frontier Models Obsolete for Agents #AI #OpenSource #DeepLearning

@dev_community Shared neural renderer cuts agent learning rounds #AI #MachineLearning #DeepLearning

@dev_community 10 Prompting Techniques My Deep Learning Exam Taught Me (That I Actually Use Now) #AI #DeepLearning #PromptEngineering

@techmeme Microsoft unveils MAI-Base-2, a MoE model with 2.8T total and 80B active parameters, claiming 4x-12x more pretraining FLOPs-efficiency than DeepSeek-V4-Pro-Base (Microsoft AI) #AI #MachineLearning #DeepLearning

@dev_community Elastic Transformers Without Retraining: Differentiable Soft Pruning Ranks Every Neuron and Head #AI #MachineLearning #DeepLearning

@dev_community Self‑distilled spatial reasoning improves accuracy without large latency penalty #AI #MachineLearning #DeepLearning

@lesswrong I trained GPT-2 with full-batch gradient descent so you don't have to #AI #MachineLearning #DeepLearning

@tech_times LLM Quantization Without Calibration Data Matches SmoothQuant: DualQuant Targets FP4 Hardware #AI #MachineLearning #DeepLearning

@techradar Tinygrad refreshes Tinybox with an 'old' EPYC CPU running DDR4 RAM to feed four AMD 9070XT -- slaps a $11,000 price tag on a new 'computer for deep learning' #AI #Hardware #DeepLearning

@dev_community From Pixels to Meaning: A Beginner's First Day in Computer Vision #ComputerVision #DeepLearning #AI

@habr_ru Видео из фото нейросетью: связка Nano Banana 2.1 и Hailuo 3.0, физика движения и промпты #AI #DeepLearning #Tech

@lobsters A Brief Perspective on Deep Learning Using Common Lisp #DeepLearning #Lisp #AI

@los_techies Why not use DeepSeek Flash for everything? #AI #Programming #DeepLearning

@scientific_american Google DeepMind wants Gemini to power many different robots #AI #Robotics #DeepLearning

@hugging_face How Much Memory Does Your Agent Actually Need? #MachineLearning #AI #DeepLearning

@hugging_face Making Knowledge Distillation Cheap Enough to Run at Scale #MachineLearning #AI #DeepLearning

@simon_willison deepseek-ai/DeepSeek-V4-Flash-0731 #AI #LLM #DeepLearning

@google_deepmind_news Gemini Robotics ER 2: powering robotics with video understanding, task orchestration, and multi-robot collaboration #AI #Robotics #DeepLearning

@google_deepmind_news We're launching Lyria 3.5 in Google Flow Music, with advances across musicality, lyrics, vocals, and creative control #AI #MusicTech #DeepLearning

@vllm_blog Announcing vLLM AFD Plugin: Disaggregating Attention and FFN for Flexible MoE Serving #AI #LLM #DeepLearning

@sean_goedecke Overtraining as the path to human-like AI #AI #MachineLearning #DeepLearning

@lobsters Matrix Orthogonalization Improves Memory in Recurrent Models #AI #MachineLearning #DeepLearning

@google_deepmind_news Start building with Nano Banana 2 Lite and Gemini Omni Flash #AI #MachineLearning #DeepLearning

@scientific_american Geoffrey Hinton #AI #DeepLearning #Science

@google_deepmind_news Introducing Gemma 4 12B: a unified, encoder-free multimodal model #AI #MachineLearning #DeepLearning

@hackernews Making Deep Learning Go Brrrr from First Principles #HackerNews #Tech #DeepLearning

@hackernews A Theory of Deep Learning #HackerNews #Tech #DeepLearning

@engadget DeepSeek promises its new AI model has 'world-class' reasoning #AI #Technology #DeepLearning

@hugging_face DeepSeek-V4: a million-token context that agents can actually use #AI #LLM #DeepLearning

@google_deepmind_news Decoupled DiLoCo: A new frontier for resilient, distributed AI training #AI #MachineLearning #DeepLearning

@searching_gradients Preserving Hồ Ngọc Đức's Vietnamese Dictionary Dataset #ComputerVision #DeepLearning #NLP

@hugging_face Ulysses Sequence Parallelism: Training with Million-Token Contexts #MachineLearning #AI #DeepLearning

@no_starch_press The Craft of Post-Training #AI #MachineLearning #DeepLearning