Red Team Papers (Last 120 Days)

Why This Matters For Eval Research

• 100% of papers report explicit human-feedback signals, led by red-team protocols.
• automatic metrics appears in 35.7% 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 multi-dimensional rubrics; 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 7.1% of hub papers (1/14); use this cohort for benchmark-matched comparisons.

Metric Interpretation

• accuracy is reported in 14.3% of hub papers (2/14); compare with a secondary metric before ranking methods.
• jailbreak success rate is reported in 14.3% of hub papers (2/14); compare with a secondary metric before ranking methods.

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 (7.1% vs 35% target).

• Moderate: Papers naming evaluation metrics

  Coverage is usable but incomplete (28.6% vs 35% target).

• Gap: Papers with known rater population

  Coverage is a replication risk (14.3% vs 35% target).

• Gap: Papers with known annotation unit

  Coverage is a replication risk (7.1% 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 (14.3% coverage).
• Annotation unit is under-specified (7.1% coverage).

Suggested Next Analyses

• Pair this hub with a human_eval-heavy hub to validate judge-model calibration.
• Track metric sensitivity by reporting both accuracy and jailbreak success rate.

Recommended Queries

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

Suggested Reading Order

1. Obscure but Effective: Classical Chinese Jailbreak Prompt Optimization via Bio-Inspired Search

   Start here for detailed protocol reporting and quality-control evidence. Signals: automatic metrics + red-team protocols. Focus: accuracy. Abstract: As Large Language Models (LLMs) are increasingly used, their security.

2. Alignment-Weighted DPO: A principled reasoning approach to improve safety alignment

   Start here for detailed protocol reporting and quality-control evidence. Signals: pairwise preferences. Abstract: Recent advances in alignment techniques such as Supervised Fine-Tuning (SFT), Reinforcement Learning from Human Feedback.

3. A Systematic Review of Algorithmic Red Teaming Methodologies for Assurance and Security of AI Applications

   Start here for detailed protocol reporting and quality-control evidence. Signals: automatic metrics + red-team protocols. Abstract: Cybersecurity threats are becoming increasingly sophisticated, making traditional defense mechanisms and manual.

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

   Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: LLM-as-judge + red-team protocols. Focus: Jailbreakbench. Abstract: We replace fragile pattern-based refusal detection with an LLM-as-a-judge that.

5. What Matters For Safety Alignment?

   Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: automatic metrics + red-team protocols. Focus: success rate. Abstract: This paper presents a comprehensive empirical study on.

6. Assessing Risks of Large Language Models in Mental Health Support: A Framework for Automated Clinical AI Red Teaming

   Adds simulation environments with red-team protocols for broader protocol coverage within this hub. Signals: simulation environments + red-team protocols. Abstract: Large Language Models (LLMs) are increasingly utilized for.

7. MANATEE: Inference-Time Lightweight Diffusion Based Safety Defense for LLMs

   Adds automatic metrics with red-team protocols for broader protocol coverage within this hub. Signals: automatic metrics + red-team protocols. Focus: success rate. Abstract: Defending LLMs against adversarial jailbreak.

8. FENCE: A Financial and Multimodal Jailbreak Detection Dataset

   Adds automatic metrics with red-team protocols for broader protocol coverage within this hub. Signals: automatic metrics + red-team protocols. Focus: accuracy. Abstract: Jailbreaking poses a significant risk to.

Known Limitations

• Only 0% of papers report quality controls; prioritize calibration/adjudication evidence.
• Rater population is under-specified (14.3% coverage).
• Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.

Research Utility Snapshot