Causal Modeling of Adverse Pregnancy Outcomes via Adaptive LLM Proposals
Authors: Kavimayil P. Komarasamy, Saurabh Mathur, Ameet Soni, David M. Haas, Kristian Kersting, Sriraam Natarajan
Organizations: The University of Texas at Dallas · German Research Center for AI (DFKI) · Technical University of Darmstadt · Swarthmore College · Indiana University School of Medicine · Hessian Center for Artificial Intelligence (hessian.ai)
Adverse Pregnancy Outcomes (APOs) such as preterm birth and gestational diabetes can have long-term consequences for both the mother and child, yet an understanding of their causes remains elusive. Causal discovery in this domain is especially challenging due to a paucity of data and incomplete domain knowledge. As a result, pure data-driven methods fail, and Large Language Model (LLM) outputs remain inconsistent or contradictory. We introduce a neurosymbolic framework for generating plausible causal hypotheses that iteratively combines the broad prior knowledge of LLMs with empirical scoring on data. Our method treats the LLM as an adaptive proposal distribution, generating hypotheses that are scored against empirical data; the resulting high-scoring graphs are then used to update the LLM's context, steering subsequent generations toward more promising regions of the hypothesis space. We evaluate our approach on a real-world clinical dataset for modeling APOs and their risk factors, comparing our results against an expert-constructed causal graph. Our method recovers all expert-validated edges and identifies additional plausible causal relations not previously listed by experts, potentially providing new insights for targeted interventions.
Figures & tables
Fig. 1 : The clara Framework . Our framework addresses causal discovery in data-sparse domains via a Generate-Evaluate-Update loop. It uses a pretrained LLM to generate diverse causal hypotheses, which are evaluated using a small clinical dataset. These scored graphs are used to identify patterns that consistently appear in top-scoring graphs and update the sampling distribution to shift mass towards more promising regions.
Variable
%
Race
Non-Hispanic Asian
4.2
Non-Hispanic Black
11.2
Non-Hispanic White
67.8
Hispanic
12.3
Other
4.4
TABLE I : Baseline characteristics and pregnancy outcomes. Variables and their frequencies in our dataset. Row colors indicate variable type: demographic , family history , pre-existing conditions , lifestyle factors , pregnancy outcomes . All risk factors were measured at the start of pregnancy.
Fig. 2 : Expert-elicited causal graph (left) and additional edges discovered by clara (right) for the obstetrics domain. Node color indicates variable type: demographic , family history , pre-existing conditions , lifestyle factors , pregnancy outcomes .
ALARM
nuMoM2b
Method
SHD ↓
SID ↓
Recall ↑
Precision ↑
SHD ↓
SID ↓
Recall ↑
Precision ↑
PC
35
314
0.5
0.6
34
83
0.1
0.3
FCI
46
368
0.1
0.5
32
79
0.0
0.0
MIMIC
66.2±10.4
473.0±15.4
0.0
0.1
37.6±3.1
73.6±2.5
0.1
0.2
LLM: GPT-5.2
LLM (One-shot)
48.2±2.7
205.7±18.0
0.6
0.4
29.2±2.7
8.8±4.4
1.0
0.5
TABLE II : Comparative evaluation of clara and baselines on ALARM and nuMoM2b domains.
Fig. 3 : Sensitivity of Structural Intervention Distance (SID) to noise and data size. SID of PC, FCI, and CLARA across dataset sizes (3k, 6k, 9k) at noise levels 0.2, 0.4, 0.6. Each panel corresponds to a fixed noise level. PC and FCI are deterministic and reported as single-run values. clara SID values are the mean ± standard deviation across five independent runs (shown as error bars) using GPT-5.2 with K=7, T=5. Lower SID indicates fewer errors in predicted interventional distributions, i.e., better recovery of causal structure.
Causal discovery seeks to uncover causal relations from data, typically represented as causal graphs, and is essential for predicting the effects of interventions. While expert knowledge is required to construct principled causal graphs, many statistical methods have been proposed to leverage observational data with varying formal guarantees. Causal Assumption-based Argumentation (ABA) is a framework that uses symbolic reasoning to ensure correspondence between input constraints and output graphs, while offering a principled way to combine data and expertise. We explore the use of large language models (LLMs) as imperfect experts for Causal ABA, eliciting semantic structural priors from variable names and descriptions and integrating them with conditional-independence evidence. Experiments on standard benchmarks and semantically grounded synthetic graphs demonstrate state-of-the-art performance, and we additionally introduce an evaluation protocol to mitigate memorisation bias when assessing LLMs for causal discovery.
Zihao Li, Fabrizio Russo
Department of Computing, Imperial College London, UK
Large language models (LLMs) are trained on enormous amounts of data and encode knowledge in their parameters. We propose a pipeline to elicit causal relationships from LLMs. Specifically, (i) we sample many documents from LLMs on a given topic, (ii) we extract an event list from from each document, (iii) we group events that appear across documents into canonical events, (iv) we construct a binary indicator vector for each document over canonical events, and (v) we estimate candidate causal graphs using causal discovery methods. Our approach does not guarantee real-world causality. Rather, it provides a framework for presenting the set of causal hypotheses that LLMs can plausibly assume, as an inspectable set of variables and candidate graphs.
Causal discovery is a cornerstone of scientific reasoning, yet whether large language models can perform it reliably remains an open question. Recent benchmarks show that even fine-tuned models plateau on simple causal graphs and degrade as complexity grows, but why they fail has not been established. We prove the failure is fundamental: supervised fine-tuning, direct preference optimization, and in-context learning all produce predictors that cannot distinguish between causal graphs generating similar observational data, and any attempt to do so requires the model's internal representations to grow unboundedly, violating the very conditions under which these methods work. We formalize this as a kernel obstruction theorem, establishing that the limitation is intrinsic to the learning paradigm, not any particular model or dataset. We propose Agentic Causal Bayesian Optimization (A-CBO), wherein a frozen language model serves as an interventional oracle answering targeted queries about intervention effects, while an external Bayesian loop concentrates beliefs over candidate graphs in logarithmically many rounds. Because the decision operates outside the space where the obstruction applies, A-CBO provably converges while the underlying model remains unchanged. On Corr2Cause, A-CBO matches fine-tuned baselines without any training. On Extended Corr2Cause, a new benchmark scaling to 24 variables with 18K test samples, A-CBO significantly outperforms both fine-tuning and preference optimization, with the advantage growing
Amartya Roy, Sonali Parbhoo
SIRE, IIT Delhi and Robert Bosch GmbH, India · Imperial College London