AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Conformal Reliability: A New Evaluation Metric for Conditional Generation

Yachen Gao, Xinwei Sun, Yikai Wang, Ye Shi, Jingya Wang, Jianfeng Feng, Yanwei Fu

ICML 2026 · Image generation › Autoregressive and latent generation · approach: Synthetic training data from generators · 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
calibrate jointly in a latent space instead of using directional quantile regression aloneCReL prediction set · nonlinear synthetic, alpha=0.10area of prediction set: 287.4 -> 232.7Improved
calibrate in latent space rather than output space (Feldman et al.)CReL prediction set · nonlinear synthetic, alpha=0.10area of prediction set: 234.5 -> 232.7No clear effect
use CReL instead of PCP for prediction setconformal calibration · nonlinear synthetic, alpha=0.1area of prediction set: 854.24 -> 232.70Improved
calibrate in latent space with linear programming instead of grid discretization in output spaceconformal calibration · calibration runtime at r=12runtime (s): 122,795.0 -> 63.54Improved
replace average metric with CReL worst-case reliability scoreimage-to-text ranking (CLIP-SIM) · MS-COCO, alpha=0.1model ranking: BLIP-base: last by CLIP (0.23304) to first by CReL-CLIP (0.00701)Mixed
replace average metric with CReL worst-case reliability scoretext-to-image ranking (DINO-SIM) · MS-COCO, alpha=0.1model ranking: SD3-M: highest average DINO 0.46151 to lowest CReL-DINO -0.14804Mixed
run projected gradient descent from several random starting points (ablation)reliability score optimization · image-to-text BLIP-base CLIP-SIM, alpha=0.1std over 10 runs: std 0.00027Improved
reduce VAE reconstruction quality (large reconstruction error) (ablation)latent generative model · coverage sensitivity analysiscoverageHurt

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