AI Paper Atlas

Research map / Optimization / Optimizers

The Double Dilemma in Multi-Task Radiology Report Generation: A Gradient Dynamics Analysis and Solution

Erjian Zhang, Yatong Hao, Liejun Wang, Zhiqing Guo

ICML 2026 · Optimization › Optimizers · approach: Surrogate losses and robust solvers · 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
replace linear scalarization with CAME-GradREVTAF · MIMIC-CXR testCE average: 0.607 -> 0.626Improved
replace linear scalarization with CAME-GradDCL · MIMIC-CXR testCE average: 0.249 -> 0.301Improved
replace linear scalarization with CAME-GradPromptMRG · MIMIC-CXR testB-1: 0.396 -> 0.396No clear effect
replace linear scalarization with CAME-GradDDaTR · IU X-Ray test (zero-shot)CE average: 0.282 -> 0.283No clear effect
replace linear scalarization with CAME-GradWCL · IU X-Ray test (supervised)CE average: 0.493 -> 0.517Improved
add conflict-averse direction rectification (S1) before magnitude enhancement (ablation)CAME-Grad on PromptMRG · MIMIC-CXR testF1: 0.476 -> 0.491Improved
add magnitude-enhanced energy injection (S2) (ablation)CAME-Grad on DDaTR · MIMIC-CXR testF1: 0.425 -> 0.438Improved
add adaptive gradient fusion (S3) (ablation)CAME-Grad on DDaTR · MIMIC-CXR testprecision: 0.425 -> 0.432Improved
add adaptive gradient fusion (S3) (ablation)CAME-Grad on PromptMRG · MIMIC-CXR testrecall: 0.531 -> 0.530No clear effect
use CAME-Grad instead of CAGrad on REVTAFREVTAF · MIMIC-CXR testCE average: 0.615 -> 0.626Improved
use MMPareto instead of linear scalarizationREVTAF · MIMIC-CXR testCE average: 0.607 -> 0.076Hurt
use GradNorm instead of linear scalarizationREVTAF · MIMIC-CXR testCE average: 0.607 -> 0.604No clear effect

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