Unverified50% confidenceFactExact time
The Survprompt framework converts structured patient covariates into free-text clinical vignettes, which are then fed to a pre-trained LLM to predict survival in a zero-shot manner.
1
Sources
50%
Confidence
Long-term
Relevance
10/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe LLM's symptom extraction process is based on a standard vocabulary written by clinicians, ensuring structured and predictable output.61% similarUnverifiedFeature ablation showed that the LLM's prioritization of clinical variables is highly similar to that of specialized survival models.60% similarUnverifiedThe arXiv study 'Large Language Models are Approximate Survival Estimators' found that frontier LLMs can provide survival prognosis estimates under zero-shot conditions that approach those of specialized models.59% similarUnverifiedParsec introduces a small language model (SLM) specifically trained for token compression, which preprocesses content before it enters the main model.59% similarVerifiedGLM uses an autoregressive blank-filling training objective, randomly masking text segments and requiring the model to autoregressively restore them.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/964028API
curl https://kongchang.com/api/v1/knowledge/claims/964028MCP
get_claim(id=964028)