Research map / Probabilistic and scientific modeling / Molecules and proteins
Beyond Accuracy: Latent Perturbations for Cognitive-Aware Diagnosis
Yuting Yan, Yinghao Fu, Wendi Ren, Haozhou Gao, Shuang Li
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use DMAE latent perturbation instead of REVISE | counterfactual generation · Gitelman | RMSE: 5.40±0.89 -> 1.93±0.76 | Improved |
| use DMAE counterfactuals instead of CFVAE | counterfactual generation · CCMS | RMSE: 0.79±0.01 -> 0.14±0.08 | Improved |
| use full loss set instead of the ablated variant (loss terms removed) (ablation) | DMAE counterfactuals · Gitelman | RMSE: 4.85±3.27 -> 1.93±0.76 | Improved |
| use full loss set instead of the ablated variant (loss terms removed) (ablation) | DMAE counterfactuals · IHPRF3 | RMSE: 0.47±0.12 -> 0.10±0.04 | Improved |
| use full loss set instead of the ablated variant (loss terms removed) (ablation) | DMAE counterfactuals · HCM | RMSE: 0.46±0.27 -> 0.10±0.13 | Improved |
| add AI-specific objectives (ablation) | DMAE · seven datasets | AI predictor performance | Improved |
| add human-centric objectives and mask regularization (ablation) | DMAE · seven datasets | human simulator performance | Improved |
| pool all seven datasets into one unified model | DMAE · public datasets | mean PRAUC: 0.93 -> 0.90 | Hurt |
| pool all seven datasets into one unified model | DMAE · private datasets | mean PRAUC: 0.96 -> 0.95 | Hurt |
| increase class imbalance (more common-disease controls) | DMAE classifier · Gitelman and HMS-GPA | AUC / PRAUC | No clear effect |
Closest papers by meaning
- Dynamic Decision Learning: Test-Time Evolution for Abnormality Grounding in Rare Diseases (ICML 2026 · similarity 0.85)
- CMR-RD: Long-Tailed Adaptive VLM for Explainable CMR Diagnosis (CVPR 2026 · similarity 0.84)
- RAD: Towards Trustworthy Retrieval-Augmented Multi-modal Clinical Diagnosis (NeurIPS 2025 · similarity 0.84)
- Large Language Models Are Clinical Reasoners: Reasoning-Aware Diagnosis Framework with Prompt-Generated Rationales (AAAI 2024 · similarity 0.84)
- Memorize and Rank: Elevating Large Language Models for Clinical Diagnosis Prediction (AAAI 2025 · similarity 0.83)
- CUPCase: Clinically Uncommon Patient Cases and Diagnoses Dataset (AAAI 2025 · similarity 0.83)
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.
