Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

GLiGuard: Schema-Conditioned Classification for LLM Safeguard

Urchade Zaratiana, Mary Newhauser, George Hurn-Maloney, Ash Lewis

Published

May 8, 2026

Citations

0

Trust level

Moderate

Usefulness score

65/100 (Medium)

Extraction confidence

70% (Moderate)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

May 8, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics Coding Red Team

Should you rely on this paper?

This paper has useful evaluation signal, but protocol completeness is partial; pair it with related papers before deciding implementation strategy.

Use this for comparison and orientation, not as your only source.

Best use

Secondary protocol comparison source

Use if you need

A secondary eval reference to pair with stronger protocol papers.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

No major weakness surfaced.

Human feedback signal
Detected
From extracted signals
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
65/100
Moderate-confidence candidate

Useful as a secondary reference; validate protocol details against neighboring papers.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions. However, state-of-the-art guardrail models rely on autoregressive decoders with 7B--27B parameters, reformulating what is fundamentally a classification problem as sequential text generation, a design choice that incurs high latency and scales poorly to multi-aspect evaluation. In this work, we introduce \textbf{GLiGuard}, a 0.3B-parameter schema-conditioned bidirectional encoder adapted from GLiNER2 for LLM content moderation. The key idea is to encode task definitions and label semantics directly into the input sequence as structured token schemas, enabling simultaneous evaluation of prompt safety, response safety, refusal detection, 14 fine-grained harm categories, and 11 jailbreak strategies in a single non-autoregressive forward pass. This schema-conditioned design lets supported task and label blocks be composed directly in the input schema at inference time. Across nine established safety benchmarks, GLiGuard achieves F1 scores competitive with 7B--27B decoder-based guards despite being 23--90$\times$ smaller, while delivering up to 16$\times$ higher throughput and 17$\times$ lower latency. These results suggest that compact bidirectional encoders can approach the accuracy of much larger guard models while drastically reducing inference cost. Code and models are available at https://github.com/fastino-ai/GLiGuard.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

strong

Red Team

Directly usable for protocol triage.

"Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions."

Evaluation Modes

strong

Automatic Metrics

Includes extracted eval setup.

"Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions."

Benchmarks / Datasets

missing

Not extracted

No benchmark anchors detected.

"Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions."

Reported Metrics

strong

Accuracy, F1, Inference cost

Useful for evaluation criteria comparison.

"These results suggest that compact bidirectional encoders can approach the accuracy of much larger guard models while drastically reducing inference cost."

Benchmarks and datasets

No benchmark or dataset names were extracted from the available abstract.

Reported metrics

accuracyf1inference cost
Human feedback details
Uses human feedback
Yes
Feedback types
Red Team
Rater population
Not reported
Expertise required
Coding
Evaluation details
Evaluation modes
Automatic Metrics
Agentic eval
None
Quality controls
Not reported
Evidence quality
Moderate
Use this page as
Secondary protocol comparison source

Research brief

Metadata summary

Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions.
  • However, state-of-the-art guardrail models rely on autoregressive decoders with 7B--27B parameters, reformulating what is fundamentally a classification problem as sequential text generation, a design choice that incurs high latency and scales poorly to multi-aspect evaluation.
  • In this work, we introduce \textbf{GLiGuard}, a 0.3B-parameter schema-conditioned bidirectional encoder adapted from GLiNER2 for LLM content moderation.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Validate inferred eval signals (Automatic metrics) against the full paper.
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

Accuracy

Contribution summary

  • Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions.
  • In this work, we introduce GLiGuard, a 0.3B-parameter schema-conditioned bidirectional encoder adapted from GLiNER2 for LLM content moderation.
  • Across nine established safety benchmarks, GLiGuard achieves F1 scores competitive with 7B--27B decoder-based guards despite being 23--90\times smaller, while delivering up to 16\times higher throughput and 17\times lower latency.

Why it matters for eval

  • Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions.
  • Across nine established safety benchmarks, GLiGuard achieves F1 scores competitive with 7B--27B decoder-based guards despite being 23--90\times smaller, while delivering up to 16\times higher throughput and 17\times lower latency.

Researcher checklist

  • Human feedback protocol is explicit

    Detected: Red Team

  • Evaluation mode is explicit

    Detected: Automatic Metrics

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Metric reporting is present

    Detected: accuracy, f1, inference cost