Skip to content

Test Report

This test has the following configuration:

data provider model dataclass temperature prompt_file active

Test Results

f1_macro f1_micro cost_summary 

Detailed Results

Result for letter05

Model did not produce a valid response!

Result for letter06

Model did not produce a valid response!

Result for letter08

Model did not produce a valid response!

Result for letter10

Model did not produce a valid response!

Result for letter11

Model did not produce a valid response!

Result for letter12

View raw result from qwen/qwen3-vl-8b-thinking

Category Ground Truth Prediction TP FP FN
send_date 1926-08-13 1926-08-13 1 0 0
sender_persons Groschupf-Jaeger, Louis
Ritter-Dreier, Fritz
Grosshöpf
J.A.Ritter
1 1 1
receiver_persons Vischer-von Planta, Max Dr. Max Vischer 1 0 0
Name Alternate Names
Groschupf-Jaeger, Louis Groschopf
Groschupf
Herr Groschupf
Herrn Groschupf
Ritter-Dreier, Fritz Fritz Ritter
Herr Fritz Ritter
Herr Ritter
Herrn Fritz Ritter
J.A. Ritter
J.A.Ritter
Ritter
Vischer-von Planta, Max Dr Max VISCHER
Dr. M. Vischer - von Planta
Dr. M. Vischer-von Planta
Dr. M. Vischer-von Planta
Dr. Max Fischer
Dr. Max Vischer
Dr. Vischer - von Planta
Dr.M. Vischer - von Planta
Herr Dr. M. Vischer-von Planta
Herr Dr. Max Vischer
Herr Dr. Vischer
Herr Dr.Vischer
Herrn Dr. Max Vischer
Herrn Dr. Vischer
Herrn Dr.Max Vischer
Herrn Präsident Dr. Max Vischer
Max Vischer
Max Vischer-von Planta
V.
Vischer
praesident vischer

Rules

inferred_from_function: False

inferred_from_correspondence: False

skip_signatures: True

skip_non_signatures: False

Letter images

letter12_p1.jpg

letter12_p2.jpg

letter12_p3.jpg

Result for letter15

Model did not produce a valid response!

Result for letter16

Model did not produce a valid response!

Result for letter18

Model did not produce a valid response!

Result for letter20

Model did not produce a valid response!

Result for letter21

Model did not produce a valid response!

Result for letter24

Model did not produce a valid response!

Result for letter26

Model did not produce a valid response!

Result for letter28

Model did not produce a valid response!

Result for letter32

Model did not produce a valid response!

Result for letter33

Model did not produce a valid response!

Result for letter36

Model did not produce a valid response!

Result for letter38

View raw result from qwen/qwen3-vl-8b-thinking

Category Ground Truth Prediction TP FP FN
send_date 1928-05-24 1938-05-24 0 1 1
sender_persons None None 0 0 0
receiver_persons Stauffacher, Werner Werner Stauffacher 1 0 0
Name Alternate Names
Stauffacher, Werner Direktor Werner Stauffacher
Herr Stauffacher
Herr Werner Stauffacher
Herren Stauffacher
Herrn Direktor Werner Stauffacher
Herrn W. Stauffacher
Stauffacher
W. Stauffacher
Werner Stauffacher

Rules

inferred_from_function: False

inferred_from_correspondence: False

skip_signatures: True

skip_non_signatures: False

Letter images

letter38_p1.jpg

letter38_p2.jpg

Result for letter46

Model did not produce a valid response!

Result for letter49

Model did not produce a valid response!

Result for letter51

Model did not produce a valid response!

Result for letter59

Model did not produce a valid response!

Result for letter61

Model did not produce a valid response!

Result for letter63

Model did not produce a valid response!

Result for letter65

Model did not produce a valid response!