method · first seen 6 Nov 2026
input filtering guardrail
Screening user prompts before they reach a language model to block harmful requests.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Jev as an LLM Safety Guardrail: An Empirical Study of Input Filtering, Selective Output Review, and Over-RefusalPapers · 6 Nov 2026 · Olivia Cognetti, Bhaskar Krishnamachari
Latest sources
- Jev as an LLM Safety Guardrail: An Empirical Study of Input Filtering, Selective Output Review, and Over-RefusalPapers · 6 Nov 2026 · Olivia Cognetti, Bhaskar Krishnamachari