BB
BenchBoard

Decision cockpit for pathology foundation-model benchmarking, built to move from broad coverage scans to task-level evidence.

← back to BenchBoard·all tasks

Task Evidence

HANCOCK · primary_tumor_site (Hancock__primary_tumor_site)

Tissue / Specimen Context / Tissue/Specimen Origin & Sampling Context

Inspect benchmark evidence for one task at a time: who leads, how stable the split-level results are, and which neighbouring tasks are likely to transfer intuition.

clsL1 · Tissue / Specimen ContextL2 · Tissue/Specimen Origin & Sampling Contexthead/neckH&EAnchor evidence

Task metadata

Fields sourced from benchmark_main. Missing values are shown as —.

cls
Identity
task_key
Hancock__primary_tumor_site
dataset_name
HANCOCK
task_name
primary_tumor_site
task_type
cls
Taxonomy
task_family_l1
Tissue / Specimen Context
task_family_l2
Tissue/Specimen Origin & Sampling Context
label_structure
multiclass
num_classes
4
labels
{"0": "Oral Cavity", "1": "Hypopharynx", "2": "Larynx", "3": "Oropharynx"}
Biology
organ
head/neck
stain
H&E
specimen_type
resection
timepoint
diagnosis
disease
头颈部癌
Scale & Labels
patient_n
693
slide_n
701
label_source
pathology
label_visibility
indirect
Result Evidence
result_tier
Anchor
result_quality_score
90
result_flags
crowded_head
result_tier_reason
Stable split behavior: median std 0.027.Clear model separation: significant-win density 0.421.Difficulty sits in a usable range: best mean 0.927.Many contenders sit near the best score: 20.2% within 0.01.
loading…

Similar tasks

Ranked by shared L2 subtype first, then L1 family, organ, type and label structure.