Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

REDACT: A Systematically Controlled Multilingual Benchmark for Personal Information Detection

Guneesh Vats, Anubha Agrawal, Shikha Singhal, Ajita Dash +4 more

Published

Jun 18, 2026

Citations

0

Trust level

Low

Usefulness score

37/100 (Low)

Extraction confidence

45% (Low)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Jun 18, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics Coding Llm As Judge Multilingual

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

The available metadata is too thin to trust this as a primary source.

Human feedback signal
Not explicit
Not explicit in abstract metadata
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
37/100
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

Benchmark infrastructure for personally identifiable information (PII) detection remains limited: existing corpora cover few entity types, use ad hoc generation conditions, and do not show which surface conditions cause detector failures. We present REDACT, a systematically controlled multilingual PII benchmark with 13,427 records, 324,078 entity annotations, 51 entity types, 4,127 surface-form patterns, and 25 languages across 9 scripts. A strength-2 covering-array sampler controls nine generation axes: domain, format, difficulty, length, density, code-switching, language, adjacency, and co-occurrence. Three entity-level metadata fields (disclosure status, disclosure form, and a GDPR-aligned sensitivity tier) enable stratified evaluation beyond aggregate or per-type F1. From the full benchmark, we evaluate five detectors (Presidio, GLiNER, the OpenAI Privacy Filter, GPT-4.1, and Claude Sonnet 4.6) on a locked, language-stratified sample of 1,000 records. Aggregate F1 masks an architecture-dependent failure structure: the rule-based detector performs poorly on the highest-stakes data, including HIGH-sensitivity categories (recall 0.07) and non-verbatim disclosure forms, while the LLM detectors remain more robust, with the HIGH tier as their strongest sensitivity slice. A three-model reference-free LLM-as-judge assessment corroborates that sensitivity-tier assignment is the task's hardest axis. We release the benchmark, schema, prompts, and stratified evaluation harness.

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.

"Benchmark infrastructure for personally identifiable information (PII) detection remains limited: existing corpora cover few entity types, use ad hoc generation conditions, and do not show which surface conditions cause detector failures."

Evaluation Modes

partial

Llm As Judge, Automatic Metrics

Includes extracted eval setup.

"Benchmark infrastructure for personally identifiable information (PII) detection remains limited: existing corpora cover few entity types, use ad hoc generation conditions, and do not show which surface conditions cause detector failures."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Benchmark infrastructure for personally identifiable information (PII) detection remains limited: existing corpora cover few entity types, use ad hoc generation conditions, and do not show which surface conditions cause detector failures."

Benchmarks / Datasets

missing

Not extracted

No benchmark anchors detected.

"Benchmark infrastructure for personally identifiable information (PII) detection remains limited: existing corpora cover few entity types, use ad hoc generation conditions, and do not show which surface conditions cause detector failures."

Reported Metrics

partial

F1, Recall

Useful for evaluation criteria comparison.

"Aggregate F1 masks an architecture-dependent failure structure: the rule-based detector performs poorly on the highest-stakes data, including HIGH-sensitivity categories (recall 0.07) and non-verbatim disclosure forms, while the LLM detectors remain more robust, with the HIGH tier as their strongest sensitivity slice."

Benchmarks and datasets

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

Reported metrics

f1recall
Human feedback details
Uses human feedback
No
Feedback types
None
Rater population
Not reported
Expertise required
Coding, Multilingual
Evaluation details
Evaluation modes
Llm As Judge, Automatic Metrics
Agentic eval
None
Quality controls
Not reported
Evidence quality
Low
Use this page as
Background context only

Research brief

Metadata summary

Benchmark infrastructure for personally identifiable information (PII) detection remains limited: existing corpora cover few entity types, use ad hoc generation conditions, and do not show which surface conditions cause detector failures.

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

Key takeaways

  • Benchmark infrastructure for personally identifiable information (PII) detection remains limited: existing corpora cover few entity types, use ad hoc generation conditions, and do not show which surface conditions cause detector failures.
  • We present REDACT, a systematically controlled multilingual PII benchmark with 13,427 records, 324,078 entity annotations, 51 entity types, 4,127 surface-form patterns, and 25 languages across 9 scripts.
  • A strength-2 covering-array sampler controls nine generation axes: domain, format, difficulty, length, density, code-switching, language, adjacency, and co-occurrence.

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

F1

Contribution summary

  • We present REDACT, a systematically controlled multilingual PII benchmark with 13,427 records, 324,078 entity annotations, 51 entity types, 4,127 surface-form patterns, and 25 languages across 9 scripts.
  • Three entity-level metadata fields (disclosure status, disclosure form, and a GDPR-aligned sensitivity tier) enable stratified evaluation beyond aggregate or per-type F1.
  • From the full benchmark, we evaluate five detectors (Presidio, GLiNER, the OpenAI Privacy Filter, GPT-4.1, and Claude Sonnet 4.6) on a locked, language-stratified sample of 1,000 records.

Why it matters for eval

  • We present REDACT, a systematically controlled multilingual PII benchmark with 13,427 records, 324,078 entity annotations, 51 entity types, 4,127 surface-form patterns, and 25 languages across 9 scripts.
  • From the full benchmark, we evaluate five detectors (Presidio, GLiNER, the OpenAI Privacy Filter, GPT-4.1, and Claude Sonnet 4.6) on a locked, language-stratified sample of 1,000 records.

Researcher checklist

  • Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Evaluation mode is explicit

    Detected: Llm As Judge, 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: f1, recall