Compartmentalised Agentic Reasoning for Clinical NLI
This addresses the need for safer, auditable reasoning in clinical NLP, though it is incremental as it builds on existing LLM methods with a novel decomposition approach.
The paper tackled the problem of improving structured reasoning in clinical natural language inference by introducing CARENLI, a compartmentalized agentic system that separates knowledge access from inference, resulting in up to 42-point fidelity improvements, such as 98.0% in Causal Attribution and 81.2% in Risk State Abstraction.
A common assumption holds that scaling data and parameters yields increasingly structured, generalisable internal representations. We interrogate this assumption in clinical natural language inference (NLI) by adopting a benchmark decomposed into four reasoning families, Causal Attribution, Compositional Grounding, Epistemic Verification, and Risk State Abstraction, and introducing CARENLI, a Compartmentalised Agentic Reasoning for Clinical NLI that separates knowledge access from principled inference. CARENLI routes each premise, statement pair to a family specific solver and enforces auditable procedures via a planner, verifier, and refiner. Across four LLMs, CARENLI improves fidelity by up to 42 points, reaching 98.0% in Causal Attribution and 81.2% in Risk State Abstraction. Verifiers flag violations with near-ceiling reliability, while refiners correct a substantial share of epistemic errors. Remaining failures cluster in routing, identifying family classification as the main bottleneck. These results show that LLMs often retain relevant facts but default to heuristics when inference is underspecified, a dissociation CARENLI makes explicit while offering a framework for safer, auditable reasoning.