Skip to content
← Back to explorer

HFEPX Hub

General + Red Team Papers

Updated from current HFEPX corpus (Apr 12, 2026). 43 papers are grouped in this hub page.

Read Full Context

Updated from current HFEPX corpus (Apr 12, 2026). 43 papers are grouped in this hub page. Common evaluation modes: Automatic Metrics, Llm As Judge. Most common rater population: Domain Experts. Common annotation unit: Trajectory. Frequently cited benchmark: Jailbreakbench. Common metric signal: jailbreak success rate. Use this page to compare protocol setup, judge behavior, and labeling design decisions before running new eval experiments. Newest paper in this set is from Apr 8, 2026.

Papers: 43 Last published: Apr 8, 2026 Global RSS Tag RSS
GeneralRed Team

Researcher Quick Triage

This hub is best used for protocol triage and replication planning from abstract-level evidence. Quality band: Developing .

All Sampled Papers (43) Replication-Ready Only (3)

High-Signal Coverage

100.0%

43 / 43 sampled papers are not low-signal flagged.

Replication-Ready Set

3

Benchmark + metric + eval mode explicitly present.

Judge/Human Comparability

0

Papers containing both `human_eval` and `llm_as_judge`.

  • 3 papers are replication-ready (benchmark + metric + explicit evaluation mode).
  • 0 papers support judge-vs-human agreement analysis.
  • 0 papers report explicit quality controls (calibration/adjudication/IAA).

Primary action: Start with the top 2 papers in “Start Here”, then validate assumptions in the protocol matrix.

Currently showing only replication-ready papers in ranking and matrix sections (3 papers).

Need evaluators for this research workflow?

Post a Job →

Why This Matters For Eval Research

  • 100% of papers report explicit human-feedback signals, led by red-team protocols.
  • automatic metrics appears in 34.9% of papers in this hub.
  • Jailbreakbench is a recurring benchmark anchor for cross-paper comparisons in this page.

Protocol Takeaways

  • Quality-control reporting is sparse in this slice; prioritize papers with explicit calibration or adjudication steps.
  • Rater context is mostly domain experts, and annotation is commonly trajectory-level annotation; use this to scope replication staffing.
  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.

Benchmark Interpretation

  • Jailbreakbench appears in 2.3% of hub papers (1/43); use this cohort for benchmark-matched comparisons.
  • Reliablebench appears in 2.3% of hub papers (1/43); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • jailbreak success rate is reported in 14% of hub papers (6/43); compare with a secondary metric before ranking methods.
  • success rate is reported in 11.6% of hub papers (5/43); compare with a secondary metric before ranking methods.
Researcher Checklist (Expanded)

Researcher Checklist

  • Strong: Papers with explicit human feedback

    Coverage is strong (100% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (0% vs 30% target).

  • Gap: Papers naming benchmarks/datasets

    Coverage is a replication risk (11.6% vs 35% target).

  • Strong: Papers naming evaluation metrics

    Coverage is strong (41.9% vs 35% target).

  • Gap: Papers with known rater population

    Coverage is a replication risk (7% vs 35% target).

  • Gap: Papers with known annotation unit

    Coverage is a replication risk (7% vs 35% target).

Strengths

  • Strong human-feedback signal (100% of papers).

Known Gaps

  • Only 0% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (7% coverage).
  • Annotation unit is under-specified (7% coverage).

Suggested Next Analyses

  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.
  • Stratify by benchmark (Jailbreakbench vs Reliablebench) before comparing methods.
  • Track metric sensitivity by reporting both jailbreak success rate and success rate.
Recommended Queries (Expanded)

Recommended Queries

LLM-as-Judge Protocols Benchmark Slice: Jailbreakbench Metric Slice: jailbreak success rate Recent High-Signal Papers

Start with These 3

Use these when you need one protocol anchor, one benchmark anchor, and one recent comparison point before reading the wider hub.

Start Here (Best First 6)

Ranked for protocol completeness (human signal, benchmark + metric anchors, quality controls, and judge/human overlap).

Protocol Matrix (Top 12)

Use this to quickly compare protocol ingredients instead of scanning long prose.

Paper HF Signal Eval Modes Benchmarks Metrics QC
TraceSafe: A Systematic Assessment of LLM Guardrails on Multi-Step Tool-Calling Trajectories

Apr 8, 2026

 Yes
Red Team
Automatic Metrics Tracesafe Bench Accuracy Not Reported
SemEval-2026 Task 6: CLARITY -- Unmasking Political Question Evasions

Mar 14, 2026

 Yes
Red Team
Automatic Metrics Semeval F1 , F1 macro Not Reported
RedTeamCUA: Realistic Adversarial Testing of Computer-Use Agents in Hybrid Web-OS Environments

May 28, 2025

 Yes
Red Team
Automatic Metrics Rtc Bench Jailbreak success rate Not Reported

Protocol Diff (Top Papers)

Fast side-by-side comparison for the highest-ranked papers in this hub.

Signal TraceSafe: A Systematic Assessment of LLM Guardrail… SemEval-2026 Task 6: CLARITY -- Unmasking Political… RedTeamCUA: Realistic Adversarial Testing of Comput…
Human Feedback Red TeamRed TeamRed Team
Evaluation Modes Automatic MetricsAutomatic MetricsAutomatic Metrics
Benchmarks Tracesafe BenchSemevalRtc Bench
Metrics AccuracyF1, F1 macroJailbreak success rate
Quality Controls Not reportedNot reportedNot reported
Rater Population UnknownDomain ExpertsUnknown
Annotation Unit TrajectoryUnknownUnknown
Suggested Reading Order (Extended)

This section is intentionally expanded only when needed; use “Start Here” above for a faster pass.

Suggested Reading Order

  1. TraceSafe: A Systematic Assessment of LLM Guardrails on Multi-Step Tool-Calling Trajectories

    Start here for detailed protocol reporting and quality-control evidence. Signals: automatic metrics + red-team protocols. Focus: Tracesafe-Bench / accuracy. Abstract: As large language models (LLMs) evolve from static.

  2. Trojan-Speak: Bypassing Constitutional Classifiers with No Jailbreak Tax via Adversarial Finetuning

    Start here for detailed protocol reporting and quality-control evidence. Signals: red-team protocols. Abstract: Fine-tuning APIs offered by major AI providers create new attack surfaces where adversaries can bypass.

  3. Prompt Attack Detection with LLM-as-a-Judge and Mixture-of-Models

    Start here for detailed protocol reporting and quality-control evidence. Signals: LLM-as-judge + red-team protocols. Focus: latency. Abstract: In production, guardrails must mitigate these attacks under strict low-latency constraints,.

  4. RedTeamCUA: Realistic Adversarial Testing of Computer-Use Agents in Hybrid Web-OS Environments

    Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: automatic metrics + red-team protocols. Focus: Rtc-Bench / jailbreak success rate. Abstract: Computer-use agents (CUAs) promise to.

  5. A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness

    Include an LLM-as-judge paper to test judge design and agreement assumptions. Signals: LLM-as-judge + red-team protocols. Focus: Reliablebench. Abstract: Automated \enquote{LLM-as-a-Judge} frameworks have become the de facto standard.

  6. Refusal Steering: Fine-grained Control over LLM Refusal Behaviour for Sensitive Topics

    Include an LLM-as-judge paper to test judge design and agreement assumptions. Signals: LLM-as-judge + red-team protocols. Focus: Jailbreakbench. Abstract: We replace fragile pattern-based refusal detection with an LLM-as-a-judge.

  7. SemEval-2026 Task 6: CLARITY -- Unmasking Political Question Evasions

    Adds automatic metrics with red-team protocols for broader protocol coverage within this hub. Signals: automatic metrics + red-team protocols. Focus: Semeval / f1. Abstract: Political speakers often avoid.

  8. Red-Teaming Vision-Language-Action Models via Quality Diversity Prompt Generation for Robust Robot Policies

    Adds simulation environments with red-team protocols for broader protocol coverage within this hub. Signals: simulation environments + red-team protocols. Focus: task success. Abstract: Furthermore, results from a user.

Known Limitations

Known Limitations

  • Only 0% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (7% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.
Research Utility Snapshot (Detailed)

Research Utility Snapshot

Human Feedback Mix

  • Red Team (43)
  • Pairwise Preference (3)
  • Rubric Rating (1)

Evaluation Modes

  • Automatic Metrics (15)
  • Llm As Judge (3)
  • Simulation Env (1)

Top Benchmarks

  • Jailbreakbench (1)
  • Reliablebench (1)
  • Rtc Bench (1)
  • Semeval (1)

Top Metrics

  • Jailbreak success rate (6)
  • Success rate (5)
  • Accuracy (3)
  • Helpfulness (3)

Rater Population Mix

  • Domain Experts (3)

Quality Controls

Coverage diagnostics (sample-based): human-feedback 100.0% · benchmarks 11.6% · metrics 39.5% · quality controls 0.0%.

Top Papers

Related Hubs

Get Started

Join the #1 Platform for AI Training Talent

Where top AI builders and expert AI Trainers connect to build the future of AI.
Self-Service
Post a Job
Post your project and get a shortlist of qualified AI Trainers and Data Labelers. Hire and manage your team in the tools you already use.
Create Account & Post a Job
Managed Service
For Large Projects
Done-for-You
We recruit, onboard, and manage a dedicated team inside your tools. End-to-end operations for large or complex projects.
Learn About Managed Service
For Freelancers
Join as an AI Trainer
Find AI training and data labeling projects across platforms, all in one place. One profile, one application process, more opportunities.
Join Now