Research map / Large language models / Instruction and fine-tuning
CircuitPrint: Mechanistic Circuit Fingerprints for Large Language Models
Zhenxiong Yan, Suhang Yao, Yu Liu, Wenqiang Jin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use circuit-based fingerprint probes instead of behavioral adversarial prompts (ProFLingo, TRAP) | LLM fingerprinting · Gemma-2-2B derivatives | FSR: 0.93 vs 0.45 and 0.33 on gemma-2-2b-it | Improved |
| use CircuitPrint instead of CTCC invasive watermarking | LLM fingerprinting · Gemma-2-2B derivatives | FSR: 0.93 vs 0.95 on gemma-2-2b-it | No clear effect |
| increase fine-tuning along the training trajectory (ablation) | CircuitPrint · Gemma-2-2B fine-tuning checkpoints | FSR: initial drop then stabilization | Mixed |
| reduce the victim's merge ratio under Task Arithmetic with DARE (ablation) | CircuitPrint · model merging | FSR | Hurt |
| raise sampling temperature above 1.0 (ablation) | CircuitPrint verification · Gemma-2-2B | TPR | Hurt |
| vary top-p between 0.55 and 1.0 (ablation) | CircuitPrint verification · Gemma-2-2B | TPR | No clear effect |
| replace direct supernode intervention with adversarial GCG prefix (ablation) | CircuitPrint probe · Gemma-2-2B and Llama-3.2-1B | collapse token agreement | No clear effect |
| intervene on random nodes instead of supernodes (ablation) | circuit interventions · derivatives of Gemma-2-2B and Llama-3.2-1B | collapse token agreement: above 85% for supernodes | Hurt |
| test on strict and relaxed near-clone models trained from scratch (ablation) | CircuitPrint · Pythia-410M | mean FSR: 0.85 -> 0.07 (strict), 0.05 (relaxed) | Hurt |
| increase suppression strength of target supernodes (ablation) | derivative models · collapse token agreement | agreement ratio | Improved |
Closest papers by meaning
- Fingerprinting Inference Systems of Large Language Models (NeurIPS 2026 · similarity 0.87)
- Hypothesis Testing the Circuit Hypothesis in LLMs (NeurIPS 2024 · similarity 0.84)
- SIF: Semantically In-Distribution Fingerprints for Large Vision-Language Models (CVPR 2026 · similarity 0.84)
- Fingerprinting Denoising Diffusion Probabilistic Models (CVPR 2025 · similarity 0.84)
- BiShield-TEE: On the (In-)Security of Unilateral Weight Obfuscation in On-Device TEE-Shielded LLM Partition (NeurIPS 2026 · similarity 0.84)
- CoreGuard: Safeguarding Foundational Capabilities of LLMs Against Model Stealing in Edge Deployment (NeurIPS 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
