Swift Sings: MDX-Net Vocal Splits and RVC Voice Conversion On-Device with ONNX/CoreML
web.navan.dev·2d
🎧Learned Audio
Flag this post
Calligraphers and Storytellers
sibylline.dev·21h·
Discuss: Hacker News
🇯🇵Japanese Computing
Flag this post
Scientists Need a Positive Vision for AI
spectrum.ieee.org·2h
🔲Cellular Automata
Flag this post
On-Policy Distillation
thinkingmachines.ai·1d·
💻Local LLMs
Flag this post
Retro Language Models: Rebuilding Karpathy's RNN in PyTorch
gilesthomas.com·4d·
Discuss: Hacker News
🧮Kolmogorov Bounds
Flag this post
Quantifying Investor Sentiment & Dynamic Capital Allocation using Multi-Modal Bayesian Networks
dev.to·11h·
Discuss: DEV
🎛️Feed Filtering
Flag this post
AI Transcription for Students on a Budget — Say Goodbye to Per-Minute Fees
dev.to·6h·
Discuss: DEV
FLAC Verification
Flag this post
Algorithmic Choreography: Generative Sonic Landscapes via Neural Field Synthesis
dev.to·3d·
Discuss: DEV
🎧Learned Audio
Flag this post
K-DAREK: Distance Aware Error for Kurkova Kolmogorov Networks
arxiv.org·1d
🧮Kolmogorov Complexity
Flag this post
Adaptive Transformer Architecture Optimization via Hyper-parameter Exploration and Reinforcement Learning
dev.to·1d·
Discuss: DEV
Z3 Optimization
Flag this post
1-Minute Ghibli Style Transformation: Unleash Your Creative Potential with Fast Image AI
dev.to·12h·
Discuss: DEV
Bidirectional Programming
Flag this post
A beginner's guide to the Video-Utils model by Nicolascoutureau on Replicate
dev.to·1d·
Discuss: DEV
🎬Video Codecs
Flag this post
Accurate and Scalable Multimodal Pathology Retrieval via Attentive Vision-Language Alignment
arxiv.org·1d
🧮Vector Embeddings
Flag this post
RAG Pipelines Made Simple: Ballerina for AI-Powered Search
dev.to·17h·
Discuss: DEV
🤖Archive Automation
Flag this post
Evaluating LLMs on Generating Age-Appropriate Child-Like Conversations
arxiv.org·11h
📝ABNF Extensions
Flag this post
Enhanced Predictive Modeling of Brazil Nut Effect Distribution via Dynamic Network Rescaling
dev.to·3h·
Discuss: DEV
🔲Cellular Automata
Flag this post
Code-enabled language models can outperform reasoning models on diverse tasks
arxiv.org·2d
🧠Intelligence Compression
Flag this post