Topic: 35
10 stories found
Yesterday
A Generative Grammar Underlying the Voynich Manuscript, the Pastiche Hypothesis: Evidence from Large Language Models
Researchers have used large language models to find evidence supporting the idea that the Voynich Manuscript uses a generative grammar similar to natural languages, challenging previous hypotheses and suggesting it may be a pastiche. This matters because deciphering or understanding the manuscript's true nature could provide insights into its creation and historical context.
Thursday, September 17, 2026
ggml/llama.cpp releases: b11026
The ggml/llama.cpp project released version b11026, which includes a model update to skip gate_up_exps if TENSOR_SKIP is set. This update is crucial for qwen35moe when MTP tensors are fused but not loaded, ensuring compatibility and functionality across different configurations.
Wednesday, September 16, 2026
ggml/llama.cpp releases: b11009
The ggml/llama.cpp project released a new version addressing issues with split states and granularity for fused QKV gemm operations, crucial for models like Qwen35. This update ensures correct handling of attention layers when using specific configurations, enhancing the model's performance and accuracy.
Monday, September 14, 2026
SynthSentry: Detecting Synthetic Data Contamination in Language Model Training Data
Physics-Informed Conformal Prediction: Embedding PDE Consistency into Distribution-Free Uncertainty Quantification for Neural Operators
Notes on gotchas while migrating 35kb preprompts from Opus to self-hosted Ollama
The article discusses challenges encountered during the migration of 35KB pre-prompts from the Opus model to a self-hosted Ollama instance, highlighting issues that could impact performance and functionality. This matters because understanding these gotchas is crucial for smooth transitions between different AI platforms.
Sunday, September 13, 2026
Tuesday, September 8, 2026
The Download: our 35 Innovators Under 35 this year
The MIT Technology Review has announced its 35 Innovators Under 35 list for 2026, highlighting young scientists and technologists driving significant advancements. This list showcases the future leaders in science and technology whose innovations could shape the tech landscape.
🌿 That's all for now. Come back tomorrow.
10 of 10 items shown. Sources: 123 days indexed.