Method / evidence and interpretation

Show the work.
Keep the limits visible.

Arcophos independently analyzes the April 2026 HealthBench Professional paper and official dataset card. Historical results remain attributed to the authors. Our contributions are the length-adjustment calculator and interpretation of selection, metrics and system conditions.

Distinguish model and system

A base model, browsing harness and product harness are separate experimental conditions.

Preserve score units

The calculator uses fractional rubric scores internally, then displays score points. It retains pre-aggregation values.

Respect source boundaries

We publish aggregate metadata and formulas without reproducing benchmark conversations or claiming an evaluation run.

What this resource cannot establish

  • The complete internal evaluation implementation is not publicly released.
  • Difficulty-enriched task scores do not estimate routine clinical failure rates.
  • The length adjustment is an empirical evaluation rule, not a recommendation about clinical answer length.

Original analysis and reported evidence

We distinguish source facts from our interpretations. Benchmark creators are credited in each dossier. Reproduced result rows are selected historical measurements from an identified paper version and configuration. They are not new Arcophos runs, a complete leaderboard, or a claim about the latest models.

When papers, repositories, or releases disagree, the dossier records which source governs the displayed claim. The evidence JSON exposes these facts, source IDs, and version notes for inspection. Dataset access and licensing are reported separately.

Analytical tools

Coverage explorers organize published task categories with our annotations; the number of displayed entries is not a performance score. Calculator inputs are explicitly illustrative. Published scoring formulas are applied to those inputs, while sampling assumptions and aggregation boundaries remain visible. These tools cannot establish clinical efficacy.

Using the worksheet

The worksheet turns the editorial guidance into a planning brief. Its completion count records items you have addressed. It does not estimate model quality, validate safety, or certify a system. The exported file includes selected and unresolved items so the remaining work stays visible.

Corrections and updates

When a source or interpretation changes, the affected guide and its update date should change together. Send the page URL, the disputed statement, and a primary reference through Arcophos contact.