LLM StructCore: Schema-Guided Reasoning Condensation and Deterministic Compilation
This work addresses the challenge of noisy language and strict output contracts in clinical data extraction, with incremental improvements in accuracy and language-agnostic design.
The paper tackled the problem of automatically filling Case Report Forms (CRFs) from clinical notes by introducing a two-stage pipeline using Schema-Guided Reasoning and deterministic compilation, achieving macro-F1 scores of 0.6543 (EN) and 0.6905 (IT) on a dev split and 0.63 on a hidden test set.
Automatically filling Case Report Forms (CRFs) from clinical notes is challenging due to noisy language, strict output contracts, and the high cost of false positives. We describe our CL4Health 2026 submission for Dyspnea CRF filling (134 items) using a contract-driven two-stage design grounded in Schema-Guided Reasoning (SGR). The key task property is extreme sparsity: the majority of fields are unknown, and official scoring penalizes both empty values and unsupported predictions. We shift from a single-step "LLM predicts 134 fields" approach to a decomposition where (i) Stage 1 produces a stable SGR-style JSON summary with exactly 9 domain keys, and (ii) Stage 2 is a fully deterministic, 0-LLM compiler that parses the Stage 1 summary, canonicalizes item names, normalizes predictions to the official controlled vocabulary, applies evidence-gated false-positive filters, and expands the output into the required 134-item format. On the dev80 split, the best teacher configuration achieves macro-F1 0.6543 (EN) and 0.6905 (IT); on the hidden test200, the submitted English variant scores 0.63 on Codabench. The pipeline is language-agnostic: Italian results match or exceed English with no language-specific engineering.