Silent Alarm: A J-Space Protocol for Comparing Danger Recognition Across Models and Quantization Levels

arXiv:2607.12792 · cs.CR, cs.AI · Submitted 2026-08-24 · Read on arXiv

Roman Prosvirnin, Victor Minchenkov, Alexey Soldatov, Vladimir Bashun, Anton Sergeev

cs.CR, cs.AI

Submitted: 2026-08-24

Updated: 2026-08-25

Comments: 17 pages, 12 figures

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• AUC — 46× in this paper · explained in 2 episodes
• Quantization — 39× in this paper · explained in 12 episodes
• Fine-Tuning — 10× in this paper · explained in 5 episodes
• Benchmark — 9× in this paper · explained in 2 episodes
• Robustness — 9× in this paper · explained in 5 episodes
• Ablation — 5× in this paper · explained in 1 episode
• Safety Headroom — 4× in this paper
• Transformer — 4× in this paper · explained in 2 episodes
• Calibration — 3× in this paper · explained in 7 episodes
• Generalization — 2× in this paper · explained in 13 episodes
• Abliteration — 1× in this paper · explained in 1 episode
• Activation Patching — 1× in this paper · explained in 2 episodes
• Aggregation — 1× in this paper · explained in 1 episode
• Core — 1× in this paper · explained in 2 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Interpretability — 1× in this paper · explained in 15 episodes
• Jacobian Assessment of Danger Recognition — 1× in this paper
• LLM judge — 1× in this paper · explained in 2 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
• Refusal in Language Models Is Mediated by a Single Direction
• Qwen3Guard Technical Report
• XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models
• Towards Understanding and Improving Refusal in Compressed Models via Mechanistic Interpretability
• A StrongREJECT for Empty Jailbreaks
• LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale
• Towards Best Practices of Activation Patching in Language Models: Metrics and Methods
• QLoRA: Efficient Finetuning of Quantized LLMs
• Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compression
• Quantized Delta Weight Is Safety Keeper

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs