- Fixed incorrect behavior of preemptor pod when preemption of the victim takes long to complete. The preemptor pod should not be circling in scheduling cycles until preemption is finished. ([#134294](https://github.com/kubernetes/kubernetes/pull/134294), [@ania-borowiec](https://github.com/ania-borowiec)) [SIG Scheduling and Testing]

- Fixed a bug in kube-scheduler where pending pod preemption caused preemptor pods to be retried more frequently. ([#134247](https://github.com/kubernetes/kubernetes/pull/134247), [@macsko](https://github.com/macsko)) [SIG Scheduling and Testing]

### (No, really, you MUST read this before you upgrade)

 - Extended PostFilterResult with a list of Pods that the PostFilter plugin preempted.
  PostFilter plugin implementations need to be updated to return the preempted Pods (if any) in the PostFilterResult from the PostFilter method; returning nil means the plugin did not preempt any Pods at all. ([#136254](https://github.com/kubernetes/kubernetes/pull/136254), [@tosi3k](https://github.com/tosi3k)) [SIG Scheduling]

- Graduated the asynchronous preemption feature in the scheduler to beta. 
  Now the feature flag (SchedulerAsyncPreemption) is enabled by default. ([#130550](https://github.com/kubernetes/kubernetes/pull/130550), [@sanposhiho](https://github.com/sanposhiho)) [SIG Scheduling]

    *   Added
        [`tf.distribute.experimental.PreemptionCheckpointHandler`](https://www.tensorflow.org/api_docs/python/tf/distribute/experimental/PreemptionCheckpointHandler)
        to handle worker preemption/maintenance and cluster-wise consistent
        error reporting for `tf.distribute.MultiWorkerMirroredStrategy`.
        Specifically, for the type of interruption with advance notice, it