AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

COPYLENS: Towards Copyrighted Characters Infringement Detection via Copyright-Aware Prompt Learning

Yaoyu Jin, Xiaochun Yang, Hong Liu, Leixia Wang, Jian Li, Rui Ding, Bin Wang

CVPR 2026 · Large language models › Language, safety and interpretability · approach: Text watermarking

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 changedWhereMeasuredEffect
add Token Learning Rate to the meta-prompt (ablation)CopyLens · COPYCHARS-TESTCohen's Kappa: 0.27 -> 0.39Improved
add Prompt Guidance (ablation)CopyLens · COPYCHARS-TESTCohen's Kappa: 0.39 -> 0.41Improved
add Masking Mechanism (ablation)CopyLens · COPYCHARS-TESTCohen's Kappa: 0.41 -> 0.53Improved
add TLR Decay (ablation)CopyLens · COPYCHARS-TESTCohen's Kappa: 0.53 -> 0.58Improved
add Local Optimum Detection (ablation)CopyLens · COPYCHARS-TESTCohen's Kappa: 0.58 -> 0.62Improved
optimize detection prompt with CopyLens instead of CopyCat manual promptQwen2-VL-7B detector · COPYCHARS-TEST (avg)Cohen's Kappa: 0.58 -> 0.62Improved
use generic LLM prompt optimizers (OPRO, APE, IPO, LLM-OPT) instead of CopyLensQwen2-VL-7B detector · COPYCHARS-TEST (avg)Cohen's Kappa: 0.58 -> 0.19-0.27Hurt
fine-tune the detector instead of prompt optimizationQwen2-VL-7B detector · COPYCHARS-TEST (avg)Cohen's Kappa: 0.58 -> 0.54Hurt
transfer the Qwen-optimized prompt zero-shot to other detectorsCopyLens prompt · COPYCHARS-TEST, LLaVA-1.5-7BCohen's Kappa (%): 45.3 -> 49.7Improved
re-optimize the prompt for the new detectorCopyLens · COPYCHARS-TEST, LLaVA-1.5-7BCohen's Kappa (%): 49.7 -> 53.0Improved
re-optimize the prompt for GPT-4o-mini-vision instead of transferCopyLens · COPYCHARS-TEST, GPT-4o-mini-visionCohen's Kappa (%): 57.2 -> 61.4Improved
CopyLens prompt evaluated on unseen IPsCopyLens · COPYCHARS-UNSEENCohen's Kappa (%): 55.9 -> 63.3Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy