Unverified50% confidenceFactExact time
The frozen representations of pathology foundation models contain information for predicting rare joint molecular phenotypes, supporting such tasks without fine-tuning.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe study uses frozen pathology foundation model representations from UNI2-h and CONCH for feature extraction without fine-tuning.74% similarUnverifiedSuch models fundamentally predict whether a variant affects a molecular phenotype rather than directly predicting clinical phenotypes, and additional biological inference is still required to go from molecular predictions to disease associations.71% similarUnverifiedA frozen model refers to a model whose weight parameters all remain unchanged during the inference phase, with no fine-tuning, adapter training, or online learning performed.69% similarUnverifiedConformal prediction can provide theoretically guaranteed coverage rates under finite samples without relying on the underlying model type (random forests, neural networks, or other black-box models).66% similarUnverifiedWhen processing molecular tasks, Scaffold Split should be used instead of random splitting; otherwise, the model's performance will be inflated due to data leakage.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/983427API
curl https://kongchang.com/api/v1/knowledge/claims/983427MCP
get_claim(id=983427)