Models & Labs
Open ASR Leaderboard Adds Private Datasets
Hugging Face Bloghigh confidence
Why it matters
- →Incorporating private datasets reduces the risk of benchmaxxing, leading to more reliable ASR evaluations.
- →The update allows for a more comprehensive assessment of ASR models across diverse accents and speech types.
- →It balances the need for openness with the necessity of robust, real-world performance metrics.
The Open ASR Leaderboard has introduced private datasets from Appen Inc. and DataoceanAI to enhance its benchmarking process. These datasets, which include a variety of accents and speech types, are intended to prevent benchmaxxing and improve the accuracy of ASR performance evaluations. The leaderboard's average Word Error Rate (WER) will continue to be calculated using public datasets by default, but users can choose to include private datasets for a more detailed analysis. This update aims to provide a more nuanced view of ASR model performance across different conditions.
Read originalMore in Models & Labs
Models & Labsmodels
llama.cpp b10955 release addresses heap corruption
The latest release of llama.cpp, b10955, tackles a critical issue of heap corruption by disabling the ggml-cpu precompiled header and fixing CACHE_LINE_SIZE ambiguity. This update ensures consistent CACHE_LINE_SIZE values across C++ kernels and C work-buffer sizing code, preventing heap-buffer-overflow and subsequent crashes. By restoring the natural include order and removing the std::hardware_destructive_interference_size branch, the update makes the value deterministic and include-order independent. This release is a technical fix that stabilizes the runtime environment for developers using llama.cpp.
llama.cpp Releases
Models & Labsmodels
llama.cpp b10956 release enhances SYCL backend
The latest llama.cpp release, b10956, introduces significant improvements to the SYCL backend, particularly for handling large k values in TOP_K operations. By implementing a radix select method, the update allows for efficient GPU-resident processing, avoiding previous limitations that forced operations to fall back to the CPU. This change enhances performance, especially in scenarios requiring large k values, such as qwen4exp's sparse-attention indexer. The update ensures that operations are more efficient and scalable, providing a notable boost in processing speed without regressing any measured shapes.