Latest Slime Could Not Continue From Checkpoint

 File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load
    raise pickle.UnpicklingError(_get_wo_message(str(e))) from None
_pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint.
        (1) In PyTorch 2.6, we changed the default value of the weights_only argument in torch.load from False to True. Re-running torch.load with weights_only set to False will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
        (2) Alternatively, to load with weights_only=True please check the recommended steps in the following error message.
        WeightsUnpickler error: Unsupported global: GLOBAL slime.utils.eval_config.EvalDatasetConfig was not an allowed global by default. Please use torch.serialization.add_safe_globals([slime.utils.eval_config.EvalDatasetConfig]) or the torch.serialization.safe_globals([slime.utils.eval_config.EvalDatasetConfig]) context manager to allowlist this global if you trust this class/function.

Whenever I want to restart from a past checkpoint. According to this error message, this is caused by this PR: https://github.com/THUDM/slime/pull/585.

Start with the checkpoint-loading path that reaches torch.serialization.load and review PR 585 for the change associated with the reported behavior. Reproduce restarting from a past checkpoint with the current setup, then verify that checkpoint loading completes successfully without the reported UnpicklingError.