SatQuery
Evaluation archive

We wanted to know
whether it actually works.

A benchmark here is not a number. It is a dataset, a protocol, a split, a set of leakage controls, a measurement, and a statement about what has and has not been verified. Experiments that fail are recorded with the same care as the ones that succeed.

01

Benchmark timeline

config_hash 78f1e3700da15aa1 ยท last verification 2026-09-25
02

Experiments

โ€”
03

Measurement

Measured: calibration (artifacts/calibration_v001.json) โ€” other plots illustrative
04

Reproducibility

A result without an environment is a rumour
Evidence notice

The experiment records above are the real per-specialist measurements, each carrying its dataset, split, protocol, source artifact and evidence state. The one frozen, VERIFIED benchmark number is the LEVIR-CD-256 change head (pooled IoU 0.8122) and the verified live deployment; the grounding, fusion, change-VQA and calibration records are measured but UNVERIFIED (two protocols or an open metric ruling). The calibration reliability plot in section 03 is a measured result, transcribed from artifacts/calibration_v001.json (change_vqa, Val, n=16441); the remaining plots in section 03 are illustrative, not measured results. No system-level end-to-end benchmark exists; the unified benchmark adapters and the router test remain NOT RUN. A number here is only as strong as the state attached to it.