AI safety and alignment · MLD-0864
Overrefusal
Declining a request that should be allowed under the intended task or policy.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Reject a benign educational question.
Original caveat
The allowed/forbidden boundary must be defined.
Context and aliases
Level: Advanced. Stages: Training · Evaluation · Research · Governance.
Tags: Safety · RL · SLM · LLM.
Aliases: False refusal.