# Large Language Models are Approximate Survival Estimators

arXiv: https://arxiv.org/abs/2609.38181 · HTML: https://mlsanity.com/paper/2609.38181/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 45 references that are in the map.
- First posted on arXiv 4 Aug 2026.

## Builds on (most cited first, 20 of 20)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [TabLLM: Few-shot Classification of Tabular Data with Large Language Models](https://mlsanity.com/paper/2210.10723/) (2022)
- [Not All Language Model Features Are One-Dimensionally Linear](https://mlsanity.com/paper/2405.14860/) (2024)
- [Deep Learning for Survival Analysis: A Review](https://mlsanity.com/paper/2305.14961/) (2023)
- [Effective Ways to Build and Evaluate Individual Survival Distributions](https://mlsanity.com/paper/1811.11347/) (2018)
- [Towards a clinically accessible radiology foundation model: open-access and lightweight, with automated evaluation](https://mlsanity.com/paper/2403.08002/) (2024)
- [Evaluating Mathematical Reasoning of Large Language Models: A Focus on Error Identification and Correction](https://mlsanity.com/paper/2406.00755/) (2024)
- [Language Models are Few-shot Learners for Prognostic Prediction](https://mlsanity.com/paper/2302.12692/) (2023)
- [Tx-LLM: A Large Language Model for Therapeutics](https://mlsanity.com/paper/2406.06316/) (2024)
- [Characterizing Delusional Spirals through Human-LLM Chat Logs](https://mlsanity.com/paper/2603.16567/) (2026)
- [Transfer Learning for Clinical Time Series Analysis using Deep Neural Networks](https://mlsanity.com/paper/1904.00655/) (2019)
- [Can LLMs Master Math? Investigating Large Language Models on Math Stack Exchange](https://mlsanity.com/paper/2404.00344/) (2024)
- [Stop ordering machine learning algorithms by their explainability! A user-centered investigation of performance and explainability](https://mlsanity.com/paper/2206.10610/) (2022)
- [Understanding LLM Embeddings for Regression](https://mlsanity.com/paper/2411.14708/) (2024)
- [SurvivalGAN: Generating Time-to-Event Data for Survival Analysis](https://mlsanity.com/paper/2302.12749/) (2023)
- [An Effective Meaningful Way to Evaluate Survival Models](https://mlsanity.com/paper/2306.01196/) (2023)
- [When accurate prediction models yield harmful self-fulfilling prophecies](https://mlsanity.com/paper/2312.01210/) (2023)
- [Decoding-based Regression](https://mlsanity.com/paper/2501.19383/) (2025)
- [Universal Abstraction: Harnessing Frontier Models to Structure Real-World Data at Scale](https://mlsanity.com/paper/2502.00943/) (2025)

## Abstract (arXiv)

Survival analysis estimates time-to-event outcomes from patient covariates and is widely used for medical risk assessment. Patients seeking prognostic information after a diagnosis may turn to large language models (LLMs), now readily accessible through consumer applications. However, whether LLMs can provide accurate survival predictions has not been rigorously evaluated. We introduce Survprompt, a framework that converts structured patient covariates into free-text clinical vignettes and prompts pre-trained LLMs to predict survival zero-shot. We benchmark Survprompt against conventional survival models, including random survival forests (RSF), across two multi-institutional pan-cancer cohorts: the publicly available MSK-CHORD cohort and a newly curated cohort from the Providence St. Joseph Health Network constructed using an LLM-based medical abstraction framework. We report censored mean absolute error (cMAE) and concordance index (c-index) and conduct feature ablations to identify variables influencing LLM predictions. Frontier LLMs achieved surprisingly competitive cMAE for individual survival times. For example, GPT-5.6-Sol achieved cMAE within 10% of state-of-the-art RSF models specifically trained for survival prediction for several cancer types and lower cMAE than RSF for prostate cancer in MSK-CHORD. Feature ablations revealed that LLMs prioritized clinical variables similarly to specialized survival models. However, LLMs showed inconsistent accuracy across cancer types and institutions and poorly discriminated between high- and low-risk patients (lower c-index). Zero-shot LLMs can generate surprisingly accurate prognostic estimates without specialized training, but their variable performance across cancer types and institutions remains an important limitation for clinical use.