GLiNER Guard: Unified Encoder Family for Production LLM Safety and Privacy
Abstract
GLiNER Guard (GLiGuard) presents a unified encoder approach for safety moderation and PII detection that achieves high throughput and competitive accuracy under latency constraints.
Production LLM systems require both safety moderation and PII detection under strict latency and cost constraints. This creates a trade-off: autoregressive moderators are accurate but expensive, while lightweight encoders are faster but less capable. We present GLiNER Guard (GLiGuard), a unified encoder that performs safety classification and PII detection in a single forward pass, simplifying safety pipelines. We introduce three variants: compact uni- and bi-encoders (145-147M) for high-throughput serving, and GLiGuard Omni (209M) for stronger moderation quality. Under dynamic batching on a single A100, the compact model reaches 193 requests/sec with P99 latency below 1s, achieving 1.6x higher throughput than GLiNER2. Omni remains competitive with much larger moderators on public safety benchmarks. We also release PII-Bench, a span-level benchmark for evaluating PII detection in end-to-end pipelines. Overall, encoder-based guardrails offer a practical low-cost alternative for always-on moderation. Models and benchmarks are released on HuggingFace.
Get this paper in your agent:
hf papers read 2605.05277 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 3
Datasets citing this paper 0
No dataset linking this paper