Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
A developer argues that multimodal models in production are bottlenecked by sampling decisions—frame rate, chunk length, resolution, and crop—rather than model capability. These preprocessing defaults silently determine …
AI systems cannot forget or forgive due to their engineering: weights retain training data permanently, context windows switch between perfect recall and total erasure, retrieval stores lack temporal decay, and logs pres…
Researchers developed an AI pipeline that digitizes paper ECGs via smartphone photos, achieving 95.51% accuracy in detecting myocardial infarction. The system runs on CPUs in under 30 seconds, enabling remote healthcare …
Researchers have developed quantum Wasserstein generative adversarial networks (GANs) that generate full-resolution MNIST images without dimensionality reduction or patchwork models, achieving state-of-the-art results wi…
Researchers introduced Single-rollout Asynchronous Optimization (SAO), a new reinforcement learning method for large language models that uses single-rollout sampling and double-side token-level clipping to improve stabi…
Efficient Long-hOrizon (ELO) learning, a new learned optimizer, surpasses traditional models like AdamW and Muon across diverse tasks including language modeling and image classification, requiring less than 7 H100 GPU-h…
Researchers introduced negative squaring, a pre-tilting technique that adjusts weights against expected rounding errors before quantization, achieving 77% error reduction in a toy recurrent network with 3-bit quantizatio…
Researchers introduced a bifidelity method that uses generative models to efficiently quantify uncertainties in parameter estimates, reducing computational costs without sacrificing accuracy. The approach was tested on n…
A developer built a Double-Tower Gated Transformer Network (GTN) to generate profitable trading signals from high-frequency EUR/USD forex data. The system uses a custom cost matrix and weighted cross-entropy loss to comb…
A developer reviews a study on comparing machine learning models in small molecule drug discovery, arguing that the credibility of the field would improve if journals banned the word 'outperform'. The developer criticize…
The U.S. Department of Energy's ARPA-E awarded a $2.7 million grant to an Iowa State University-led team to use machine learning and innovative synthesis to discover new magnetic materials that can outperform neodymium-i…
InductWave, a wavelet-based inductive embedding method for logical query answering on knowledge graphs, outperforms existing models while using fewer message-passing layers, offering resource efficiency and scalability f…
Researchers introduced STAGformer, a novel AI model that improves bike-sharing demand forecasting by integrating spatio-temporal agent graph attention, reducing computational complexity to O(NT). Tested on NYC Citi-Bike …
Researchers have developed MiLSD, a sub-megabyte line segment detector for embedded systems, achieving a leap in performance on the ShanghaiTech Wireframe dataset from 10.6 to 24.1 sAP10 under a one-megabyte memory const…
Text-to-Speech (TTS) technology converts written text into spoken audio through stages including text normalization, phonetic transcription, prosody generation, and waveform synthesis. Machine learning, particularly deep…
A new AI-powered tool lets users type trading strategies in plain English, backtest them against 20 million data entries including historical market regimes like the COVID crash, and see results with costs included, all …
Researchers developed STST-JEPA, a self-supervised transformer model that predicts brain age from EEG data with a mean absolute error of 3.06 years, outperforming previous methods. The model, pretrained on over 47,000 se…
Researchers developed musicPIIrate, an AI tool using deep learning to infer personal attributes like age, country, and lifestyle habits from public music playlists, raising privacy concerns. In response, they created Jam…
Researchers have developed a method to compress key-value caches in AI models using contractive iterated-map codes, achieving 36-54x memory reduction on GPT-2 with a perplexity cost of 11-15%. The technique also enables …
Researchers introduced CompDiff, a generative model that improves fairness in medical imaging by addressing demographic imbalances through a Hierarchical Conditioner Network. In tests on chest X-rays and fundus images, C…