AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

DOBF: A Deobfuscation Pre-Training Objective for Programming Languages

Marie-anne Lachaux, Baptiste Roziere, Marc Szafraniec, Guillaume Lample

NeurIPS 2021 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervised speech and brain decoding · 17 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
Pre-train with DOBF+DAE instead of DOBF alone (ablation)DOBF · CodeXGLUE Code Summarization JavaBLEU: 19.0546 -> 19.3646Improved
Pre-train with DOBF+DAE instead of DOBF alone (ablation)DOBF · CodeXGLUE Code Summarization PythonBLEU: 18.2399 -> 18.5767Improved
Pre-train with DOBF+DAE instead of DOBF alone (ablation)DOBF · CodeXGLUE NL Code Search PythonMRR: 0.383 -> 0.3966Improved
Pre-train with DOBF+DAE instead of DOBF alone (ablation)DOBF · TransCoder Python→Java, beam size 10CA@1: 44.12 -> 47.27Improved
Pre-train with DOBF+DAE instead of DOBF alone (ablation)DOBF · TransCoder Java→Python, beam size 10CA@1: 39.96 -> 42.36Improved
Pre-train with DOBF+DAE instead of DOBF alone (ablation)DOBF · CodeXGLUE Clone Detection JavaF1: 95.87 -> 95.82Hurt
Initialize DOBF with MLM instead of training from scratch (ablation)DOBF · Identifier Name Proposal, eval p_obf=0accuracy: 56.3272 -> 67.58465Improved
Train with p_obf=0.5 instead of p_obf=0 (ablation)DOBF · Identifier Name Proposal, eval p_obf=0accuracy: 56.3272 -> 61.06085Improved
Train with p_obf=1 instead of p_obf=0.5 (ablation)DOBF · Identifier Name Proposal, eval p_obf=0accuracy: 61.06085 -> 18.12775Hurt
Train with p_obf=1 instead of p_obf=0.5 (ablation)DOBF · Full Deobfuscation, eval p_obf=1accuracy: 41.80665 -> 45.56205Improved
Initialize DOBF with MLM instead of training from scratch (ablation)DOBF · Full Deobfuscation, eval p_obf=1accuracy: 45.56205 -> 49.7472Improved
Use DOBF pre-training rather than MLM pre-trainingCodeXGLUE NL Code Search Python · CodeXGLUE NL Code Search PythonMRR: 0.3076 -> 0.383Improved

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