Unverified50% confidenceOpinionExact time
Large models possess cross-knowledge integration and natural language reasoning capabilities, but whether they can reliably complete hypothesis revision and avoid "anchoring" on early erroneous judgments remains a key difficulty in evaluating their clinical usability.
1
Sources
50%
Confidence
Long-term
Relevance
10/25/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAfter current mainstream large language models form incorrect premises early in their reasoning, subsequent steps tend to rationalize those premises to maintain consistency rather than backtracking to correct them.78% similarUnverifiedLarge models perform relatively well in articulating treatment principles and organizing diagnostic reasoning, because traditional Chinese medicine literature contains abundant structured knowledge mappings between syndrome types and treatment principles, allowing models to reproduce reasoning chains through pattern matching.77% similarUnverified模型的平台风格适配依赖世界知识泛化而非硬编码规则，因此存在推断偏差的可能，仍需人工复核76% similarUnverified思维链推理的效果随模型规模呈现涌现性，在较小模型上几乎无效，在足够大的模型上才能显著提升推理准确率76% similarUnverified语言模型对工具输出有一定判断力，外部知识与模型内部信念的偏差程度越大模型越不相信76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1097674API
curl https://kongchang.com/api/v1/knowledge/claims/1097674MCP
get_claim(id=1097674)