AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

PrivCode++: Latent-Conditioned Differentially Private Code Generation for Comprehensive Guarantees

Zheng Liu, Chen GONG, Terry Yue Zhuo, Zhou Yang, Kecen Li, Wenlong Meng, Xinwen Hou, Yu Liu, Xiaochen Li

ICML 2026 · Federated learning and privacy › Differential privacy · approach: Differentially private synthetic data · 0 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
add latent conditioning (PrivLC) vs unconditional generation (ablation)PrivCode++ · Instruct-HumanEval (avg of 4 models)average pass@1: 31.1 -> 52.0Improved
add latent alignment between code and prompt latents (ablation)PrivCode++ · Complete-MBPP (avg of 4 models)average pass@1: 47.7 -> 68.7Improved
raise latent dimension from 256 to 768 (ablation)PrivCode++ · Instruct HumanEval (avg of 4 models)average pass@1: 43.4 -> 52.0Improved
raise latent dimension from 768 to 2048 (ablation)PrivCode++ · Instruct HumanEval (avg of 4 models)average pass@1: 52.0 -> 47.9Hurt
use latent-conditioned synthesis instead of prompt-conditioned synthesis with public promptsPrivCode++ vs PrivCode · BigCodeBench Complete Full (Qwen2.5-Coder-7B)pass@1: 27.9 -> 43.3Improved
use latent-conditioned synthesis instead of PrivCode relaxed-privacyPrivCode++ vs PrivCode · HumanEval Instruct (DS-Coder-6.7B)pass@1: 56.1 -> 41.5Hurt
use PrivCode++ instead of DP instruction fine-tuning (DPFT)PrivCode++ vs DPFT · HumanEval Instruct (Qwen2.5-Coder-7B)pass@1: 62.2 -> 68.3Improved
condition on prompt embeddings alone instead of code latents (ablation)PC-PromptEmb vs PrivCode++ · HumanEval Instruct (Qwen2.5-Coder-7B)pass@1: 68.3 -> 54.9Hurt
use PrivCode++ latent conditioning instead of PrivCode with sensitive promptsPrivCode++ · Joint canary leakage, count=100, Qwen2.5-Coder-7Bleakage rate: 20% -> 0%Improved
increase privacy budget epsilon (ablation)PrivCode++ · HumanEval/MBPP average pass@1pass@1Improved
use public-pretrained prefix embedding (PC-PreEmb) instead of DP-trained latent (ablation)PC-PreEmb vs PrivCode++ · HumanEval Complete (Qwen2.5-Coder-7B)pass@1: 64.0 -> 51.2Hurt

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