classification.

*   `tf.data`:

    *   Added support for cross-trainer data caching in tf.data service. This
        saves computation resources when concurrent training jobs train from the
        same dataset. See
        (https://www.tensorflow.org/api_docs/python/tf/data/experimental/service#sharing_tfdata_service_with_concurrent_trainers)
        for more details.

  fix: ensure InstanceShutdownByProviderID return false for creating Azure VMs ([#104449](https://github.com/kubernetes/kubernetes/pull/104449), [@feiskyer](https://github.com/feiskyer)) [SIG Cloud Provider]
- Fixed a bug in the job controller to not mutate the shared cache ([#104130](https://github.com/kubernetes/kubernetes/pull/104130), [@adtac](https://github.com/adtac)) [SIG Apps]