Practical AI/Brief

Program-solve interface improves clinical math for 32B model, not 7B

A preprint tests whether a language model can write case-specific Python for a restricted executor instead of doing clinical arithmetic itself. The handoff helped a 32B open-weight model but not a 7B one.

BriefPublished 12 September 20261 min read1 linked source · 5 checked factsRevision 4

supercomputer cluster of compute racks in a NASA machine room — Illustration: Discover Supercomputer (GSFC) · NASA/Goddard Space Flight Center · PD licence

Researchers tested an alternative to hardcoding clinical calculators: the model writes case-specific Python that a restricted local executor runs as a deterministic solver. On MedCalc-Bench Verified (1,100 cases, 55 calculators), with formulas and gold variables supplied, the approach helped Qwen2.5-32B-AWQ (90.53% vs 83.47%, +7.05 points, 95% CI [0.47, 14.60]) but not Qwen2.5-7B (75.31% vs 72.02%, +3.29 points, 95% CI [-3.49, 10.38]).

A hand-written 22-calculator library was exact on its 440 supported cases but abstained elsewhere, scoring 40.0% overall. The authors audited the benchmark's formulas against current clinical guidelines and flagged 16 of 55 with version, use or coefficient concerns.

What the reporting says

Each line below is stated by the page named above it, and only by that page.

Stated by arXiv

The study evaluated a Program-Solve interface on MedCalc-Bench Verified, covering 1,100 cases and 55 calculators.
With formulas and gold variables supplied, handing off to the solver was not a reliable advantage at 7B (75.31% vs 72.02%, +3.29 points, 95% CI [-3.49, 10.38]).
At 32B, handing off to the solver was a reliable advantage (90.53% vs 83.47%, +7.05 points, 95% CI [0.47, 14.60]).
The hand-written 22-calculator library was exact on its 440 supported cases but abstained elsewhere, scoring 40.0% overall.
The benchmark's formulas were audited against current clinical guidelines, flagging 16 of 55 with version, use or coefficient concerns.

Our view

This approach may help larger models but does not substitute for verified formulas or reliable variable extraction.

What the reporting says: The study reports that handing off to the solver is a reliable advantage at 32B but not at 7B, and that adding an executor helps some open-weight models more than others even under matched formula, variable and note access.

Sources

arXivPublished 12 September 2026 · checked 14 September 2026

Found through monitored feeds: arXiv. A feed address is how the desk finds a story, not the evidence for it — the sources listed above are the pages the facts were taken from.

How this story was checked

This story was written from the 1 page listed above, all read on 14 September 2026.
Every figure and identifier in the text was required to appear in that source text before publication; any that did not was removed rather than checked by memory.
Of the 5 reported statements recorded for this story, 5 are each traceable to a linked page and are shown above. The rest were dropped.
Quantities traced to the source text: yes; identifiers traced: yes.
Where the reporting was silent, the text says so instead of filling the gap. This is a reported story, not a review of the source.

Program-solve interface improves clinical math for 32B model, not 7B

What the reporting says

Our view

Sources

Related reading

More from Practical AI

Autonomous AI Agents: What They Can Do and How to Manage the Risks

OpenRouter's GPT Sol Latest Alias Always Points to the Newest GPT Sol Model

Agent can study an unfamiliar environment before test time, paper finds

How this story was checked

What the reporting says

Our view

Sources

Related reading

Clinical calculator AI: code solver lifts 32B model, not 7B, preprint finds

More from Practical AI

Autonomous AI Agents: What They Can Do and How to Manage the Risks

OpenRouter's GPT Sol Latest Alias Always Points to the Newest GPT Sol Model

Agent can study an unfamiliar environment before test time, paper finds

How this story was checked