Equilibrium Matching
raywang4.github.io·1d·
Discuss: Hacker News
📊Learned Metrics
Theory of Scaling Laws for In-Context Regression: Depth, Width, Context and Time
arxiv.org·3d
🧠Machine Learning
To Augment or Not to Augment? Diagnosing Distributional Symmetry Breaking
arxiv.org·2d
🧠Machine Learning
The Phoenix of Neural Networks: Training Sparse Networks from Scratch
dev.to·3d·
Discuss: DEV
📊Quantization
Modeling Others' Minds as Code
arxiv.org·2d
🧠Intelligence Compression
Decoding Netflix's AV1 Streams: Here are 10 things I found
singhkays.com·4d·
Discuss: Hacker News
🎬AV1 Encoding
Discrete Wavelet Transform as a Facilitator for Expressive Latent Space Representation in Variational Autoencoders in Satellite Imagery
arxiv.org·3d
📊Quantization
Contrastive Retrieval Heads Improve Attention-Based Re-Ranking
arxiv.org·2d
🔍Information Retrieval
Enhanced Arabic-language cyberbullying detection: deep embedding and transformer (BERT) approaches
arxiv.org·2d
📝Text Parsing
Learning Time-Series Representations by Hierarchical Uniformity-Tolerance Latent Balancing
arxiv.org·2d
🧠Machine Learning
Fine-Tuning Masked Diffusion for Provable Self-Correction
arxiv.org·2d
💻Local LLMs
Adaptive Heterogeneous Mixtures of Normalising Flows for Robust Variational Inference
arxiv.org·2d
🧠Machine Learning
Sora 2: What is it, what can it do & how to use
dev.to·2d·
Discuss: DEV
🎬WebCodecs
Fine-Tuning Flow Matching via Maximum Likelihood Estimation of Reconstructions
arxiv.org·2d
📊Quantization
Building Ethical AI: A Comprehensive Guide to Responsible Artificial Intelligence
dev.to·2d·
Discuss: DEV
🤖AI Curation
PEL-NAS: Search Space Partitioned Architecture Prompt Co-Evolutionary LLM-driven Hardware-Aware Neural Architecture Search
arxiv.org·2d
💻Local LLMs
Sora 2: AI Video Generation with Realistic Sound
2-sora.com·2d·
Discuss: Hacker News
🎧Learned Audio
AudioMoG: Guiding Audio Generation with Mixture-of-Guidance
arxiv.org·5d
🎧Learned Audio
Show HN: TorchSystem, Event driven systems with PyTorch
github.com·1d·
Discuss: Hacker News
⚡Incremental Computation
RewardMap: Tackling Sparse Rewards in Fine-grained Visual Reasoning via Multi-Stage Reinforcement Learning
arxiv.org·2d
🔲Cellular Automata