BB
BenchBoard

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

Task Evidence

BRACS · fine_diagnosis (BRACS__slidelevel_fine)

Diagnostic Pathology / Disease Entity & Histologic Diagnosis

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 · Diagnostic PathologyL2 · Disease Entity & Histologic DiagnosisbreastH&EAnchor evidence

Task metadata

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

cls
Identity
task_key
BRACS__slidelevel_fine
dataset_name
BRACS
task_name
fine_diagnosis
task_type
cls
Taxonomy
task_family_l1
Diagnostic Pathology
task_family_l2
Disease Entity & Histologic Diagnosis
label_structure
multiclass
num_classes
7
labels
{"0": "Atypical ductal hyperplasia", "1": "Ductal carcinoma in situ", "2": "Flat epithelial atypia", "3": "Invasive carcinoma", "4": "Normal", "5": "Pathological benign", "6": "Usual ductal hyperplasia"}
Biology
organ
breast
stain
H&E
specimen_type
biopsy
timepoint
diagnosis
disease
乳腺癌
Scale & Labels
patient_n
189
slide_n
545
label_source
pathology
label_visibility
direct
Result Evidence
result_tier
Anchor
result_quality_score
92.1
result_flags
result_tier_reason
Stable split behavior: median std 0.024.Clear model separation: significant-win density 0.436.Difficulty sits in a usable range: best mean 0.872.
loading…

Similar tasks

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