Unverified50% confidenceFactExact time
The study reproduced this recipe from scratch on the MedMCQA dataset and validated it on three frozen models.
1
Sources
50%
Confidence
Long-term
Relevance
11/19/2026
First Seen
Sources
Related Entities
Related Claims
Unverified该实验的核心方法是冻结Prompt、Model config、Tool schema、Trace四个关键要素以实现可复现的回归检测66% similarUnverifiedThe study proposes a training-free six-class wafer map defect classification framework that leverages a frozen large multimodal model and achieves zero-shot classification through connected-component highlighting and rule-ordering prompts.62% similarUnverifiedIn simulation experiments over 270 grid points and under non-Gaussian seed distributions, the closed-form formula was validated within the relevant design range (from five runs per method).62% similarUnverifiedThe study adopts a preregistration-style methodology: each decision rule was frozen before running, and all reported numbers were recomputed by independent verification scripts.60% similarUnverifiedThe study is accompanied by a companion archive containing a frozen benchmark test set, evaluation audit logs, raw responses from the consensus experiments, and analysis code.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1186757API
curl https://kongchang.com/api/v1/knowledge/claims/1186757MCP
get_claim(id=1186757)