AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Veli: Unsupervised Method and Unified Benchmark for Low-Cost Air Quality Sensor Correction

Yahia Dalbah, Marcel Worring, Yen-Chia Hsu

AAAI 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Nonparametric and PAC-Bayes Bayesian learning · 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
apply Veli instead of raw LCS readingsLow-cost sensors · In-distribution UtrechtMAE: 24.77 -> 5.25Improved
apply Veli instead of raw LCS readingsLow-cost sensors · In-distribution RotterdamMAE: 21.27 -> 3.36Improved
apply Veli to already accurate sensorsLow-cost sensors · In-distribution NijmegenMAE: 2.82 -> 3.06Hurt
use Veli instead of Kalman filter denoisingKF · In-distribution UtrechtMAE: 15.95 -> 5.25Improved
fine-tune the encoder on target-region data with decoder frozen (30 epochs)Veli · Out-of-distribution TainanMAE: 8.59 -> 7.83Improved
fine-tune the encoder on target-region data with decoder frozen (30 epochs)Veli · Out-of-distribution TaichungMAE: 7.78 -> 7.65Improved
apply zero-shot Veli instead of raw LCSLow-cost sensors · Out-of-distribution TainanMAE: 14.09 -> 8.59Improved
apply KF denoising instead of raw LCSLow-cost sensors · Out-of-distribution TaichungMAE: 10.01 -> 9.98No clear effect
inject NA readings into 1 to 9 of the 10 sensors (ablation)Veli · In-distribution test sitesMAEHurt
reduce number of sensors from 10 to 3, 5 or 7 (ablation)Veli · In-distribution test sitesMAENo clear effect
set beta_y to 0 (ablation)Veli · Loss weight sensitivitytraining lossHurt
lower beta_z relative to other loss weights (ablation)Veli · Loss weight sensitivityMAEHurt

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