Detect toxic voice-chat transcripts for review
Judges: one transcribed utterance with prior two lines. Questions: choice:toxicity - none/banter/harass/threat/hate; noul:targeted_at_individual. Action: Threat and hate go to human queue within seconds; banter ignored.
STop tier. Meets A, and the payoff is major with high confidence.5.60
Key facts
- Vertical
- Games & entertainment
- Function
- Social & creator
- Status
- Idea
- Volume
- high
- Value
- major
- Risk
- high
- Evidence
- —
- Flags
- human-review
Build this with a classifier
Define a typed decision with a bounded answer, then evaluate it on examples.
{
"decision_type": "yes_no",
"question": "Does this input match the decision in “Detect toxic voice-chat transcripts for review”?",
"input": "<input to classify>",
"output": "yes | no"
}Related use cases
Cite this
Copy a link in your preferred format.