mlsanityData to 14 Oct 2026

Risk-Adjusted Harm Scoring for Automated Red Teaming for LLMs in Financial Services

arXiv 2603.10807 ↗First posted 11 Mar 2026q-fin.CP cs.AI cs.CYcited by 1

On the map: LLM safety and security → Safety alignment, refusal and guardrails

Abstract (arXiv)

Existing LLM safety evaluations rely on binary attack-success rates and domain-agnostic taxonomies, leaving regulated Banking, Financial Services, and Insurance (BFSI) deployments exposed to failures elicited through legally or professionally plausible framing. We introduce RAHS (Risk-Adjusted Harm Score), a risk-sensitive metric jointly capturing disclosure severity, disclaimer mitigation, and inter-judge agreement, and FinRedTeamBench, a 989-prompt benchmark spanning seven BFSI risk areas and 34 sub-categories mapped to regulatory frameworks. Evaluation uses an ensemble of three heterogeneous LLM judges, validated against human experts, and an adaptive multi-turn red-teaming pipeline. On nine open-weight models, RAHS preserves separation under near-ceiling ASR, ranking is stable under hyperparameter sweeps, and multi-turn pressure drives not only more jailbreaks but more operationally severe disclosures, exposing failure modes that single-turn, domain-agnostic evaluations cannot reveal.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security. Builds on 21 of its 40 references that are in the map. First posted on arXiv 11 Mar 2026.

Builds on

21 of 40 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024SG-Bench: Evaluating LLM Safety Generalization Across Diverse Tasks and Prompt Types29 Oct

Other references · 20

  1. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 3,104
  2. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,429
  3. 2023Jailbreaking Black Box Large Language Models in Twenty Queriescited by 1,420
  4. 2022Red Teaming Language Models with Language Modelscited by 1,026
  5. 2023DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Modelscited by 518
  6. 2024Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attackcited by 390
  7. 2023Large Language Models in Finance: A Surveycited by 226
  8. 2024WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Modelscited by 223
  9. 2025X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agentscited by 125
  10. 2024BetterBench: Assessing AI Benchmarks, Uncovering Issues, and Establishing Best Practicescited by 114
  11. 2023LLM360: Towards Fully Transparent Open-Source LLMscited by 104
  12. 2024A Survey of Large Language Models in Finance (FinLLMs)cited by 79
8 more
  1. 2024When LLM Meets DRL: Advancing Jailbreaking Efficiency via DRL-guided Searchcited by 47
  2. 2024Automated Red Teaming with GOAT: the Generative Offensive Agent Testercited by 45
  3. 2024RedAgent: Red Teaming Large Language Models with Context-aware Autonomous Language Agentcited by 44
  4. 2024h4rm3l: A language for Composable Jailbreak Attack Synthesiscited by 24
  5. 2024BELLS: A Framework Towards Future Proof Benchmarks for the Evaluation of LLM Safeguardscited by 14
  6. 2025JADES: A Universal Framework for Jailbreak Assessment via Decompositional Scoringcited by 14
  7. 2024LeakAgent: RL-based Red-teaming Agent for LLM Privacy Leakagecited by 13
  8. 2025Learning to Conceal Risk: Controllable Multi-turn Red Teaming for LLMs in the Financial Domaincited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026FinRT: Distilling Adaptive Red-Teaming Strategies into Reusable Adversarial Generators in Consumer Financecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026FinRED: An Expert-Guided Benchmark Generation and Evaluation Framework for Financial LLM Red-Teaming
  2. 2026FinSafetyBench: Evaluating LLM Safety in Real-World Financial Scenarios
  3. 2026LLM Judges Inconsistently Disagree Across Safety Criteria and Harm Categories
  4. 2025LLM-Safety Evaluations Lack Robustness
  5. 2026When No Benchmark Exists: Validating Comparative LLM Safety Scoring Without Ground-Truth Labels
  6. 2025TRIDENT: Benchmarking LLM Safety in Finance, Medicine, and Law
  7. 2026EvalSafetyGap: A Hybrid Survey and Conceptual Framework for LLM Evaluation-Safety Failures
  8. 2026aiXamine: Unified Black-Box Evaluation of Cross-Dimensional Trade-offs in LLM Safety, Security, and Privacy