AI Paper Atlas

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

MSDS: A Large-Scale Chinese Signature and Token Digit String Dataset for Handwriting Verification

Peirong Zhang, Jiajia Jiang, Yuliang Liu, Lianwen Jin

NeurIPS 2022 · Large language models › Language, safety and interpretability · approach: Data and evaluation methodology · 3 citations (OpenAlex)

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
Use Token Digit Strings instead of Chinese signaturesHandwriting verification systems · MSDS-TDS vs MSDS-ChS, same-user collection conditionsEERImproved
Test across sessions rather than within a single sessionHandwriting verification systems · MSDS-ChS and MSDS-TDSEERHurt
Add offline images to online time-series features (ablation)Sig2Vec verifier · MSDS-ChS, skilled forgery 4vs1EER (global/local, %): 9.03/4.97 -> 8.44/4.07Improved
Add offline images to online time-series features (ablation)Sig2Vec verifier · MSDS-ChS, random forgery 1vs1EER (global/local, %): 5.09/1.18 -> 3.54/1.53Mixed
Add offline images to online time-series features (ablation)Sig2Vec verifier · MSDS-TDS, skilled forgery 4vs1EER (global/local, %): 5.18/2.07 -> 4.95/1.88Improved
Add offline images to online time-series features (ablation)Sig2Vec verifier · MSDS-TDS, random forgery 1vs1EER (global/local, %): 1.76/0.28 -> 1.62/0.53Mixed
Use DsDTW rather than the other evaluated systemsOnline handwriting verification · Skilled-forgery scenarios on MSDS-ChS and MSDS-TDSEERImproved
Use DTW rather than the other evaluated systemsOnline handwriting verification · Random-forgery scenarios on MSDS-ChS and MSDS-TDSEERImproved
Train DsDTW on DeepSignDB rather than MSDS-ChSDsDTW · MSDS-ChS, random forgery 4vs1EER (global/local, %): 0.84/0.11 -> 0.41/0.06Improved
Train on DeepSignDB and test on Chinese signaturesDsDTW · MSDS-ChS, skilled forgery 4vs1EER (global/local, %): 5.91/2.90 -> 10.60/5.78Hurt
Use four or three genuine templates rather than oneHandwriting verification systems · MSDS-ChS and MSDS-TDSEERImproved

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