Add Dynamic Filter

def filter_function(args, samples, **kwargs):
    # return DynamicFilterOutput or bool

Preferred return type:

from slime.rollout.filter_hub.base_types import DynamicFilterOutput

return DynamicFilterOutput(keep=True, reason=None)

Buffer filter (called in slime/rollout/data_source.py):

def buffer_filter(args, rollout_id, buffer, num_samples):
    return selected_groups

Rollout sample filter:

def rollout_sample_filter(args, samples):
    # modify sample.remove_sample in-place where needed

All-samples process:

def process_all_samples(args, all_samples, data_source):
    ...

Step 3: Preserve Group Structure

• Keep list[list[Sample]] structure intact where required.
• Do not flatten sample groups unless downstream path expects flattened samples.
• For sample removal, prefer sample.remove_sample=True over deleting objects.

Step 4: Wire and Validate

Example wiring:

--dynamic-sampling-filter-path slime.rollout.filter_hub.dynamic_sampling_filters.check_reward_nonzero_std
--buffer-filter-path <module>.buffer_filter
--rollout-sample-filter-path <module>.rollout_sample_filter
--rollout-all-samples-process-path <module>.process_all_samples

Step 5: Measure Side Effects

• Check final sample count remains aligned with rollout_batch_size expectations.
• Verify drop reasons are surfaced in rollout metrics when dynamic filter is used.
• Validate training still receives valid loss masks/rewards after filtering.

Common Mistakes

• Returning wrong container type for buffer filter
• Dropping samples by deletion instead of mask flagging
• Losing sample-group alignment in group-RM setup
• Adding expensive logic in hot filtering paths

Reference Locations

• Dynamic filter types: slime/rollout/filter_hub/base_types.py
• Dynamic filter example: slime/rollout/filter_hub/dynamic_sampling_filters.py
• Rollout generation hook points: slime/rollout/sglang_rollout.py
• Buffer filter hook point: slime/rollout/data_source.py