Show HN: Videolangua – end-to-end video translate and subtitle/ dub
videolangua.com·19h·
Discuss: Hacker News
📦MKV Containers
Show HN: A Fun Flow Field-Based Image Displacement Mapping Experiment
labs.galleri5.com·1d·
Discuss: Hacker News
📐Projective Geometry
I spent weeks understanding Netflix's recommendation system - here's what I learned (Matrix Factorization breakdown + working code)
beyondit.blog·1d
🎛️Feed Filtering
Podcast Episode: Separating AI Hope from AI Hype
eff.org·1d
🤖AI Curation
How Amazon scaled Rufus by building multi-node inference using AWS Trainium chips and vLLM
aws.amazon.com·22h
⚡Incremental Computation
ParallelSearch: Train your LLMs to Decompose Query and Search Sub-queries in Parallel with Reinforcement Learning
arxiv.org·12h
🔍Information Retrieval
AST-n: A Fast Sampling Approach for Low-Dose CT Reconstruction using Diffusion Models
arxiv.org·12h
✅Lossless Bounds
SkySplat: Generalizable 3D Gaussian Splatting from Multi-Temporal Sparse Satellite Images
arxiv.org·12h
🏺Computational Archaeology
Deep and diverse population synthesis for multi-person households using generative models
arxiv.org·12h
🌍Cultural Algorithms
What is PCA? A Simplified Explanation
dev.to·1d·
Discuss: DEV
📐Vector Dimensionality
Arm neural technology to adds AI acceleration to Arm GPUs, enables “Neural Super Sampling” for lower bandwidth/higher FPS
cnx-software.com·1d
🎯Emulator Accuracy
Building a Neural Network is Easy. The Hard Part is Making It Think
dev.to·2d·
Discuss: DEV
🎧Learned Audio
SOFA: Deep Learning Framework for Simulating and Optimizing Atrial Fibrillation Ablation
arxiv.org·2d
🧠Neural Codecs
A Physics-Driven Neural Network with Parameter Embedding for Generating Quantitative MR Maps from Weighted Images
arxiv.org·2d
🧠Machine Learning
When and how can inexact generative models still sample from the data manifold?
arxiv.org·2d
🌀Differential Geometry
Sound Signal Synthesis with Auxiliary Classifier GAN, COVID-19 cough as an example
arxiv.org·1d
🧠Machine Learning
Dynamic Recurrent Neural Network Fusion for Granular Material Flow Optimization
dev.to·4d·
Discuss: DEV
🌊Stream Processing
TAR-TVG: Enhancing VLMs with Timestamp Anchor-Constrained Reasoning for Temporal Video Grounding
arxiv.org·2d
⏱️SMPTE Timecode
Adaptive Optimization of Time-Sensitive Network (TSN) Protocols via Reinforcement Learning and Dynamic Signal Shaping
dev.to·15h·
Discuss: DEV
📡Network Calculus
Bridging Semantic Logic Gaps: A Cognition-Inspired Multimodal Boundary-Preserving Network for Image Manipulation Localization
arxiv.org·2d
🤖Advanced OCR