This article from LightOn AI researchers describes a Straight-Through Estimator-based regularization technique that improves hierarchical pooling in ColBERT-style retrieval models.…
This article introduces Atom2.7m, a 2.74-million-parameter language model that reaches 69.24% accuracy on arithmetic benchmarks through an arithmetic-aware numeric representation…
Published July 8, 2026, this article surveys how frontier AI labs implement knowledge distillation through three approaches: training smaller student…
PyTorch engineers described an LLM-guided autotuning method for Helion kernels that matches the performance of the existing Likelihood-Free Bayesian Optimization…
Developer Simon Willison described a technique for using coding agents more efficiently by letting a top-tier model exercise judgment about…
Hugging Face detailed improvements to the transformers library’s vLLM backend that let it match the throughput of hand-written native vLLM…
PyTorch 2.13 was released on July 8, 2026, bringing FlexAttention support to Apple Silicon (MPS) with up to 12x speedups…
Hugging Face and SkyPilot announced an integration allowing users to mount Hugging Face Buckets and Hub repositories as storage backends…
Hugging Face rolled out major updates to its Kernels project, which standardizes distribution of custom compute kernels, adding kernels as…
Hugging Face released LeRobot v0.6.0, introducing three world model policies — VLA-JEPA, LingBot-VA, and FastWAM — that learn to imagine…
PyTorch’s compiler achieves significant speedups through kernel fusion, a technique that combines dependent operations into single, efficient kernels. By fusing…
PyTorch’s test infrastructure dynamically generates tests across devices and data types at import time, which is why CI test failures…
