Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

Harm Is Behavioural, Moderation Is Lexical: How Language and Culture Decide Whom Automated Safety Systems Protect and Whom They Silence

Gabriel Lars Sabadin, Dr Nicola Harding, Sarah Kingston

Published

Oct 22, 2026

Citations

0

Trust level

Low

Usefulness score

0/100 (Low)

Extraction confidence

35% (Low)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Oct 22, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics

Should you rely on this paper?

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Use this as background context only. Do not make protocol decisions from this page alone.

Best use

Background context only

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

This paper looks adjacent to evaluation work, but not like a strong protocol reference.

Human feedback signal
Not explicit
Not explicit in abstract metadata
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
0/100
Adjacent candidate

Treat as adjacent context, not a core eval-method reference.

If you are doing eval pipeline work, start here

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

Abstract

Content warning: this paper quotes slurs and hateful statements from published test suites; the racial slur among them is masked. Automated systems decide at scale which messages are harmful, and they decide unevenly. Classifiers label African American English as offensive up to twice as often as other English, the accounts of prominent drag queens can score as more toxic than those of white nationalists, and safety behaviour that holds in English collapses in low-resource languages. We argue that over-flagging and under-protection share one cause: deciding harm from vocabulary and surface form rather than from what a message does. We organise the evidence into seven failure modes and argue, as testable hypotheses, that harmful behaviours (a request for secrecy, a demand for payment, an attempt to isolate someone) travel across languages better than the words that express them. We propose an evaluation protocol that makes the gap between the best- and worst-served language communities a headline result, and show that the naive gap is inflated under equal service and that its usual bootstrap interval is unreliable. Applied to five open systems in eleven languages, from a keyword lexicon to two current multilingual guard models, the protocol finds that none reaches native-speaker accuracy at separating harmful function from harmless form (51 of 55 system-language pairs fall short); that greater capability raises discrimination in nine languages while leaving Arabic and romanised Hindi near chance, so the gap between the best- and worst-served languages grows with capability; that one system both over-flags and under-protects the same language most; and that a language a system reads worst can look the most fairly treated when only false positives are reported.

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

missing

None explicit

No explicit feedback protocol extracted.

"Content warning: this paper quotes slurs and hateful statements from published test suites; the racial slur among them is masked."

Evaluation Modes

partial

Automatic Metrics

Includes extracted eval setup.

"Content warning: this paper quotes slurs and hateful statements from published test suites; the racial slur among them is masked."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Content warning: this paper quotes slurs and hateful statements from published test suites; the racial slur among them is masked."

Benchmarks / Datasets

missing

Not extracted

No benchmark anchors detected.

"Content warning: this paper quotes slurs and hateful statements from published test suites; the racial slur among them is masked."

Reported Metrics

partial

Accuracy

Useful for evaluation criteria comparison.

"Applied to five open systems in eleven languages, from a keyword lexicon to two current multilingual guard models, the protocol finds that none reaches native-speaker accuracy at separating harmful function from harmless form (51 of 55 system-language pairs fall short); that greater capability raises discrimination in nine languages while leaving Arabic and romanised Hindi near chance, so the gap between the best- and worst-served languages grows with capability; that one system both over-flags and under-protects the same language most; and that a language a system reads worst can look the most fairly treated when only false positives are reported."

Benchmarks and datasets

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

Reported metrics

accuracy
Human feedback details
Uses human feedback
No
Feedback types
None
Rater population
Not reported
Expertise required
Multilingual
Evaluation details
Evaluation modes
Automatic Metrics
Agentic eval
None
Quality controls
Not reported
Evidence quality
Low
Use this page as
Background context only

Research brief

Metadata summary

Content warning: this paper quotes slurs and hateful statements from published test suites; the racial slur among them is masked.

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

Key takeaways

  • Content warning: this paper quotes slurs and hateful statements from published test suites; the racial slur among them is masked.
  • Automated systems decide at scale which messages are harmful, and they decide unevenly.
  • Classifiers label African American English as offensive up to twice as often as other English, the accounts of prominent drag queens can score as more toxic than those of white nationalists, and safety behaviour that holds in English collapses in low-resource languages.

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

  • Classifiers label African American English as offensive up to twice as often as other English, the accounts of prominent drag queens can score as more toxic than those of white nationalists, and safety behaviour that holds in English…
  • We propose an evaluation protocol that makes the gap between the best- and worst-served language communities a headline result, and show that the naive gap is inflated under equal service and that its usual bootstrap interval is unreliable.
  • Applied to five open systems in eleven languages, from a keyword lexicon to two current multilingual guard models, the protocol finds that none reaches native-speaker accuracy at separating harmful function from harmless form (51 of 55…

Why it matters for eval

  • Classifiers label African American English as offensive up to twice as often as other English, the accounts of prominent drag queens can score as more toxic than those of white nationalists, and safety behaviour that holds in English…
  • We propose an evaluation protocol that makes the gap between the best- and worst-served language communities a headline result, and show that the naive gap is inflated under equal service and that its usual bootstrap interval is unreliable.

Researcher checklist

  • Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • 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