MANATEE: Inference-Time Lightweight Diffusion Based Safety Defense for LLMs
Chun Yan Ryan Kan, Tommy Tran, Vedant Yadav, Ava Cai, Kevin Zhu, Ruizhe Li, Maheep Chaudhary
Paper appears method- or tooling-adjacent to AI workflows with partial ecosystem coverage.
Defending LLMs against adversarial jailbreak attacks remains an open challenge. Existing defenses rely on binary classifiers that fail when adversarial input falls outside the learned decision boundary, and repeated fine-tuning is computationally expensive while potentially degrading model capabilities. We propose MANATEE, an inference-time defense that uses density estimation over a benign representation manifold. M ...
ANATEE learns the score function of benign hidden states and uses diffusion to project anomalous representations toward safe regions--requiring no harmful training data and no architectural modifications. Experiments across Mistral-7B-Instruct, Llama-3.1-8B-Instruct, and Gemma-2-9B-it demonstrate that MANATEE reduce Attack Success Rate by up to 100\% on certain datasets, while preserving model utility on benign inputs.
Results & Benchmarks
No concrete benchmark grounding is available yet. Treat the page as context or an implementation starting point only.
Defending LLMs against adversarial jailbreak attacks remains an open challenge.
Implementation Evidence Summary
Recommendation evidence is currently too limited for a maintained-repo choice. Use Implementation Status and Reproduction Path for a practical baseline plan.
Reproduction Risks
- Estimate is based on paper-only reproduction flow
Hardware Notes
Expect multi-day setup/compute for meaningful reproduction based on current guidance.
Evidence disclosure
Evidence graph: 2 refs, 1 links.
Utility signals: depth 40/100, grounding 58/100, status low.
Implementation Status
There is no verified maintained implementation yet. Use this baseline plan to decide whether to prototype now or defer.
- No direct maintained implementation was found. Use the paper PDF and citation graph to design a baseline reproduction.
- Start from this likely method family: Diffusion.
- Track assumptions and missing details in an experiment log before coding.
Reproduction readiness
Hardware requirements
- Expect multi-day setup/compute for meaningful reproduction based on current guidance.
No verified implementation available
- · No maintained repository has been identified for this paper. Check adjacent implementations or HF artifacts below.
No benchmark numbers could be verified. You will not be able to validate reproduction correctness against published numbers.
Framework baselines
- Hugging Face Diffusers training guide
Practical baseline for diffusion model reproduction.
Hugging Face artifacts
No trustworthy direct or curated related Hugging Face artifacts were found yet.
Continue with targeted Hugging Face searches derived from the paper title and method context:
Tip: start with models, then check datasets/spaces if you need evaluation data or demos.
Direct artifact matches are currently sparse. Use targeted Hugging Face searches to quickly locate candidate models, datasets, and demos.
Research context
Tasks
None detected
Methods
Diffusion
Domains
None detected
Evaluation & Human Feedback Data
Open this paper in HFEPX to review benchmark signals, evaluation modes, and human-feedback protocol context.
Open in HFEPXExplore Similar Papers
Jump to Paper2Code search queries derived from this paper's research context.
Need human evaluators for your AI research? Scale annotation with expert AI Trainers.