---
title: "Auditable Hierarchical Inference with Multimodal Evidence and Bayesian Active Inquiry"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2610.38774v1"
permalink: "https://feed.astralor.com/items/itm_4912571cac4d9b351fd07c7d"
published: "2026-10-28T19:48:34.000Z"
score: 55
channel: "tech"
tags: ["推理效率","安全","RAG","多模态","Auditable Hierarchical Inference","Multimodal Evidence","Bayesian Active Inquiry","Most"]
---

# Auditable Hierarchical Inference with Multimodal Evidence and Bayesian Active Inquiry

## 速览

Most multimodal classifiers assume that all useful evidence is available upfront, whereas clinical inquiry is sequential. We present a synthetic evaluation framework separating objective multimodal inference from active…

## 关键点

- Most multimodal classifiers assume that all useful evidence is available upfront, whereas clinical inquiry is sequential.
- We present a synthetic evaluation framework separating objective multimodal inference from active inquiry.
- An objective model maps synthetic tongue, bilateral multipressure pulse, and acoustic observations to eight syndrome families.

## 全文

Most multimodal classifiers assume that all useful evidence is available upfront, whereas clinical inquiry is sequential. We present a synthetic evaluation framework separating objective multimodal inference from active inquiry. An objective model maps synthetic tongue, bilateral multipressure pulse, and acoustic observations to eight syndrome families. A provenance-bearing Knowledge Core and training-only answer model support Bayesian narrowing over 33 fine terms using 25 typed questions, while seven safety questions remain outside the diagnostic posterior. Weak shared symptom factors limit fine-term influence on objective phenotypes, and an objective-only probe tests leakage. In a preregistered 600-case synthetic evaluation, objective family accuracy was 87.0%. Cost-sensitive expected information gain (cost-EIG) raised family accuracy to 95.5% and exact fine-term accuracy to 42.5%, versus 92.7% and 33.0% for an unoptimized fixed-order baseline. Selective family accuracy reached 96.9% at 95.3% coverage. However, cost-EIG asked more questions on average than fixed order (9.795 versus 9.352). Missing-audio routing retained 94.0% family accuracy, and a workflow invariant enforced abstention when all objective modalities were absent. Permuting fine-term answer likelihoods reduced family accuracy to 78.7%. Five of six frozen checks passed, but the question-efficiency hypothesis failed. Comparisons are descriptive because only aggregate, not paired case-level, outputs were retained. These results establish a reproducible synthetic workflow, not clinical diagnostic or treatment validity.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_4912571cac4d9b351fd07c7d