mlsanityData to 14 Oct 2026

Granite Guardian

arXiv 2412.07724 ↗First posted 10 Dec 2024cs.CLcited by 25

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

Abstract (arXiv)

We introduce the Granite Guardian models, a suite of safeguards designed to provide risk detection for prompts and responses, enabling safe and responsible use in combination with any large language model (LLM). These models offer comprehensive coverage across multiple risk dimensions, including social bias, profanity, violence, sexual content, unethical behavior, jailbreaking, and hallucination-related risks such as context relevance, groundedness, and answer relevance for retrieval-augmented generation (RAG). Trained on a unique dataset combining human annotations from diverse sources and synthetic data, Granite Guardian models address risks typically overlooked by traditional risk detection models, such as jailbreaks and RAG-specific issues. With AUC scores of 0.871 and 0.854 on harmful content and RAG-hallucination-related benchmarks respectively, Granite Guardian is the most generalizable and competitive model available in the space. Released as open-source, Granite Guardian aims to promote responsible AI development across the community. https://github.com/ibm-granite/granite-guardian

Key facts. Cited by 25 papers in the map (data through 14 Oct 2026). Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security. First posted on arXiv 10 Dec 2024.

Builds on

0 of 0 references.

No references here yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 10; 2026: 14.

25 papers.

Sort:

By area:

  1. 2026Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agentscited by 30
  2. 2025Benign Samples Matter! Fine-tuning On Outlier Benign Samples Severely Breaks Safetycited by 22
  3. 2025Large Reasoning Models Learn Better Alignment from Flawed Thinkingcited by 13
  4. 2025SafeTy Reasoning Elicitation Alignment for Multi-Turn Dialoguescited by 7
  5. 2025Disentangled Safety Adapters Enable Efficient Guardrails and Flexible Inference-Time Alignmentcited by 6
  6. 2026DARWIN: Evolving Jailbreak Adversary and Guardrail for LLM Safety Evaluation and Protectioncited by 6
  7. 2025Evaluating the Robustness of Large Language Model Safety Guardrails Against Adversarial Attackscited by 5
  8. 2026AudioGuard: Toward Comprehensive Audio Safety Protection Across Diverse Threat Modelscited by 3
  9. 2025HalluGuard: Evidence-Grounded Small Reasoning Models to Mitigate Hallucinations in Retrieval-Augmented Generationcited by 2
  10. 2025AprielGuardcited by 2
  11. 2024VerifiableFL: Verifiable Claims for Federated Learning using Exclavescited by 1
  12. 2025Identifying Features Associated with Bias Against 93 Stigmatized Groups in Language Models and Guardrail Model Safety Mitigationcited by 1
13 more
  1. 2026TWGuard: A Case Study of LLM Safety Guardrails for Localized Linguistic Contextscited by 1
  2. 2026Cognitive Firewall: A Proactive, Zero-Trust, Multi-Gate Framework for LLM Safetycited by 1
  3. 2025Dark LLMs: The Growing Threat of Unaligned AI Modelscited by 0
  4. 2025Concealment of Intent: A Game-Theoretic Analysiscited by 0
  5. 2026Answering the Wrong Question: Reasoning Trace Inversion for Abstention in LLMscited by 0
  6. 2026HyperSafe: Inference-Time Safety Recovery for Fine-Tuned Language Modelscited by 0
  7. 2026CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimizationcited by 0
  8. 2026Semalith v1.4: A Calibrated 184M Safety Classifier Achieving State-of-the-Art Prompt-Injection Detection at 44x Fewer Parameters than Llama-Guard-3-8Bcited by 0
  9. 2026Policy-as-logic for robust reasoning over rulescited by 0
  10. 2026Probing the Prefill: Detecting Code Vulnerabilities via Latent Activationscited by 0
  11. 2026Towards Mitigating Deceptive Safety Alignment in Large Reasoning Modelscited by 0
  12. 2026Towards a Unified Misuse Monitoring Benchmarkcited by 0
  13. 2026Safeguarding LLMs via Model-Agnostic Latent Safety Signals from Dark Knowledgecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM
  2. 2024Luna: An Evaluation Foundation Model to Catch Language Model Hallucinations with High Accuracy and Low Cost
  3. 2025Safety and Security Analysis of Large Language Models: Benchmarking Risk Profile and Harm Potential
  4. 2024WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs
  5. 2024Controlling Risk of Retrieval-augmented Generation: A Counterfactual Prompting Framework
  6. 2025ASTRAL: Automated Safety Testing of Large Language Models
  7. 2025Understanding Gen Alpha Digital Language: Evaluation of LLM Safety Systems for Content Moderation
  8. 2025A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy