Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

RAS: Measuring LLM Safety Through Refusal Alignment

Chang-Chieh Huang, Yan-Lun Chen, Chia-Mu Yu, Wei-Bin Lee

Published

Jun 24, 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

Jun 24, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics General 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

Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy. Although useful, output-level evaluation is expensive, sensitive to judge choice, and easily tied to fixed question banks. We propose **SafeVec**, a white-box evaluation procedure that measures safety from internal representations rather than generated answers. **SafeVec** first extracts layer-wise refusal directions from a safety-aligned reference model, then selects stable layer windows where safe and unsafe behaviors are separable, and finally scores a target model by measuring whether its hidden states align with these refusal directions under unsafe and jailbreak prompts. The resulting metric, **RAS** (**R**efusal **A**lignment **S**core), maps representation-level refusal alignment to a calibrated 0-100 safety score. Across `Llama`, `Gemma`, and `Qwen` model families, RAS separates aligned models from uncensored and abliterated variants, tracks output-level attack success rate, and is substantially faster than judge-based evaluation. These results suggest that refusal alignment provides a compact and efficient signal for white-box LLM safety evaluation.

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.

"Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy."

Evaluation Modes

strong

Automatic Metrics

Includes extracted eval setup.

"Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy."

Benchmarks / Datasets

missing

Not extracted

No benchmark anchors detected.

"Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy."

Reported Metrics

strong

Success rate, Jailbreak success rate

Useful for evaluation criteria comparison.

"Across `Llama`, `Gemma`, and `Qwen` model families, RAS separates aligned models from uncensored and abliterated variants, tracks output-level attack success rate, and is substantially faster than judge-based evaluation."

Benchmarks and datasets

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

Reported metrics

success ratejailbreak success rate
Human feedback details
Uses human feedback
Yes
Feedback types
Red Team
Rater population
Not reported
Expertise required
General
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

Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy.

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

Key takeaways

  • Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy.
  • Although useful, output-level evaluation is expensive, sensitive to judge choice, and easily tied to fixed question banks.
  • We propose **SafeVec**, a white-box evaluation procedure that measures safety from internal representations rather than generated answers.

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

Contribution summary

  • Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy.
  • Although useful, output-level evaluation is expensive, sensitive to judge choice, and easily tied to fixed question banks.
  • We propose **SafeVec**, a white-box evaluation procedure that measures safety from internal representations rather than generated answers.

Why it matters for eval

  • Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy.
  • We propose **SafeVec**, a white-box evaluation procedure that measures safety from internal representations rather than generated answers.

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: success rate, jailbreak success rate