Research map / Large language models / Agents, code and math
PASS: Probabilistic Agentic Supernet Sampling for Interpretable and Adaptive Chest X-Ray Reasoning
Yushi Feng, Junye Du, Yingying Hong, Qifan Wang, Lequan Yu
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 |
|---|---|---|---|
| add early-exit action (ablation) | PASS · CAB-E | accuracy: 88.60 -> 91.22 | Improved |
| add contrastive path-rank pretraining (ablation) | PASS · CAB-E | accuracy: 87.86 -> 91.22 | Improved |
| add expert knowledge warm-up (ablation) | PASS · CAB-E | accuracy: 88.89 -> 91.22 | Improved |
| probabilistic supernet sampling vs finetuned GPT-4o on same training data | GPT-4o (finetuned) · CAB-E | accuracy: 81.82 -> 91.22 | Improved |
| PASS vs strongest baseline MedRAX | MedRAX · CAB-E | accuracy: 89.54 -> 91.22 | Improved |
| PASS vs MedRAX on safety-critical split | MedRAX · CAB-E safety-critical split | accuracy: 89.79 -> 93.50 | Improved |
| PASS vs MedRAX hallucination rate | MedRAX · CAB-E safety-critical split | hallucination rate (%): 1.60 -> 1.60 | No clear effect |
| PASS vs LLaVA-Med on LLM-Judge score | LLaVA-Med · CAB-E | LLM-Judge score: 82.65 -> 84.28 | Improved |
| PASS vs MedRAX on SLAKE (zero-shot) | MedRAX · SLAKE | embedding similarity: 74.90 -> 79.55 | Improved |
| PASS vs CheXagent / GPT-4o finetuned on SLAKE similarity | PASS · SLAKE | embedding similarity: 81.82 (GPT-4o finetuned) vs 79.55 (PASS) | Hurt |
| PASS has higher latency than single-pass models | PASS · CAB-E | latency: LLaVA-Med 21.43 vs PASS 22.06 | Hurt |
| increase cost penalty lambda to 0.03 (ablation) | PASS · CAB-E | cost / accuracy: cost down roughly 20%, accuracy around 88% | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MedRAX — MedRAX: Medical Reasoning Agent for Chest X-ray (ICML 2025) · better on 5 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- MedRAX: Medical Reasoning Agent for Chest X-ray (ICML 2025 · similarity 0.89)
- CXReasonBench: A Benchmark for Evaluating Structured Diagnostic Reasoning in Chest X-rays (NeurIPS 2025 · similarity 0.86)
- AOR: Anatomical Ontology-Guided Reasoning for Medical Large Multimodal Model in Chest X-Ray Interpretation (NeurIPS 2025 · similarity 0.86)
- CARE: Towards Clinical Accountability in Multi-Modal Medical Reasoning with an Evidence-Grounded Agentic Framework (ICLR 2026 · similarity 0.85)
- Rethinking Visual Attribution for Chest X-ray Reasoning in Large Vision Language Models (NeurIPS 2026 · similarity 0.85)
- MedAgent-Pro: Towards Evidence-based Multi-modal Medical Diagnosis via Reasoning Agentic Workflow (ICLR 2026 · similarity 0.85)
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.
