Toward System One Clinical Intelligence: Parameter-Efficient Fine-Tuning and Counterfactual Explainability of Decision Foundation Models
Automated disease classification from patient-reported symptoms requires balancing the speed of lightweight discriminative models against the contextual understanding and latency costs of autoregressive large language models (LLMs).
ProofPaper ↗
Key points
- Decision foundation models offer an alternative by mapping representations directly to candidate classes without free-form text generation.
- We adapt Clef-Flash, a 9-billion-parameter decision model built on a Qwen3.5 backbone, for 24-class disease classification on the Symptom2Disease benchmark.
- To examine symptom-level evidence, we introduce a Leave-One-Symptom-Out (LOSO) counterfactual perturbation framework.
- These findings suggest that decision-based architectures can combine the linguistic capabilities of large foundation models with low-latency inference and symptom-level interpretability for clinical triage.
Sources (1)
- [1]Toward System One Clinical Intelligence: Parameter-Efficient Fine-Tuning and Counterfactual Explainability of Decision Foundation ModelsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 10, 04:13 PM
Automated disease classification from patient-reported symptoms requires balancing the speed of lightweight discriminative models against the contextual understanding and latency costs of autoregressive large language models (LLMs).
Decision foundation models offer an alternative by mapping representations directly to candidate classes without free-form text generation.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 10, 2026Training-Free BioNEL-RAG: Advancing Biomedical Entity Linking through Retrieval-Augmented Generation and Large Language Model
- Oct 10, 2026vllm-sr/d3
- Oct 10, 2026SAGE: Pushing 2-Bit Vector-Quantized LLMs toward Real-World Usability
- Oct 10, 2026Vela: User Behavior as Supervision Sparks LLM Intelligence
- Oct 9, 2026Microsoft's Decision-1 model enters the fast-growing AI decision model race
- Oct 9, 2026Minions: A Dynamic and Distributed Virtual GPU Cloud Built on Idle Apple Silicon Devices