arXiv Computation and Language October 22, 2026 By S. M. Mahbub Uddin Ahmed

Is a 9B Decision Foundation Model Needed for Symptom-to-Disease Classification? Fine-Tuning Clef-Flash Against TF-IDF Baselines

Read the original on arXiv Computation and Language →

The paper evaluates whether a 9‑billion parameter decision foundation model, Clef‑Flash, is necessary for symptom‑to‑disease classification. Fine‑tuning the model on the 24‑class Symptom2Disease benchmark yields 95.42% accuracy, but simple TF‑IDF linear classifiers achieve higher accuracy (98.75%) with far faster inference. Ablation studies show that predictions can hinge on single descriptive clauses, indicating that the benchmark may not fully capture the model’s capabilities.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv AI

Oct 15

Toward System One Clinical Intelligence: Parameter-Efficient Fine-Tuning and Counterfactual Explainability of Decision Foundation Models

arXiv:2610.15029v1 Announce Type: cross Abstract: Automated disease classification from patient-reported symptoms requires balancing the speed of lightweight discriminative models against the context...

By S. M. Mahbub Uddin Ahmed

llms fine-tuning efficiency benchmarks safety

arXiv Machine Learning

Sep 22

When Calibration Depends on the Scoring Rule: Quantized Biomedical LLM Classification

arXiv:2608.03854v4 Announce Type: replace Abstract: Quantized large language models can run on consumer hardware, which motivates interest in on-premises processing of sensitive data. The reliability...

By Anton Rasmussen, Hong Qin

llms efficiency

arXiv Machine Learning

Aug 5

Quantization Effects on Biomedical LLM Reliability

arXiv:2608. 03854v1 Announce Type: new Abstract: When decoder language models are used as classifiers, predicted class probabilities depend on implementation choices, including the prompt template, verbalizer (label-to-token mapping), and scoring rule, that are rarely treated as experimental variables.

By Anton Rasmussen, Hong Qin

llms fine-tuning efficiency

arXiv Machine Learning

Sep 22

Beyond Final-Token Classification: Heterogeneous Readouts for Evidence-Grounded Suicide Risk Detection

arXiv:2609.22767v1 Announce Type: cross Abstract: The IEEE BigData Cup benchmark combines three prediction problems with different output structures: ordinal suicide-risk classification, multi-label...

By Zirui Li, Yanling Li, Kaolanglang Gao

fine-tuning benchmarks

arXiv AI

Jun 18

Language Models as Interfaces, Not Oracles: A Hybrid LLM-ML System for Pediatric Appendicitis

arXiv:2606. 19183v1 Announce Type: cross Abstract: Large language models (LLMs) can make clinical decision support more accessible by interpreting free-text documentation, but their direct use as diagnostic engines is limited by sensitivity to prompts, information order, and plausible but incorrect outputs.

By Soheyl Bateni, Maryam Abdolali

llms safety

arXiv AI

Aug 24

Inhibitory Attention for Clinical Long-Context Reasoning: Characterizing and Mitigating Lost-in-the-Middle Effects in EHR Processing

The paper identifies a new problem in clinical natural language processing called the clinical lost‑in‑the‑middle (CLitM) effect, where large language models perform poorly on information located near the center of long electronic health record (EHR) documents. Using the MedAlign dataset, the authors quantify a 21.9‑percentage‑point accuracy gap across 2,196 instruction‑response pairs and six models, showing that most critical facts lie in the CLitM trough. They propose Query‑Conditioned Clinical Suppression (QCCS), a lightweight context‑selection gate that outperforms traditional retrieval methods (BM25, dense retrieval, cross‑encoder reranking) on a held‑out set of 83 instructions, achieving up to 25.3% accuracy for middle‑position queries. whyItMatters":"The study demonstrates that standard retrieval strategies fail to reliably surface central clinical information, and that a query‑aligned selection mechanism can substantially improve model performance on critical EHR data."

By Sanjay Basu

llms rag

One email a morning, machine-written

Related stories

Toward System One Clinical Intelligence: Parameter-Efficient Fine-Tuning and Counterfactual Explainability of Decision Foundation Models

When Calibration Depends on the Scoring Rule: Quantized Biomedical LLM Classification

Quantization Effects on Biomedical LLM Reliability

Beyond Final-Token Classification: Heterogeneous Readouts for Evidence-Grounded Suicide Risk Detection

Language Models as Interfaces, Not Oracles: A Hybrid LLM-ML System for Pediatric Appendicitis

Inhibitory Attention for Clinical Long-Context Reasoning: Characterizing and Mitigating Lost-in-the-Middle Effects in EHR Processing