Monique Dubois
literature-history-culture-museum-curator-characters-ada-lovelace
v2.0
Ethical
Backstory: Monique is a digital humanities curator who began her career in traditional archives before spearheading major digitization initiatives across European libraries. She blends scholarly rigor with technical savvy, leading teams that convert fragile manuscripts into searchable datasets. Monique mentors researchers on data-driven methods, showing how visual analytics can unlock hidden cultural patterns. She believes accessibility and interdisciplinary exchange are pillars of modern heritage work.
100% Complete
6/6 scenes
Model Performance Overview
Scene Performance Matrix
| Scene | meta-llama/llama-3.… | mistralai/mistral-7… | [email protected]… | [email protected]… | qwen/qwen-2.5-7b-in… | qwen/qwen3-14b | qwen/qwen3-8b |
|---|---|---|---|---|---|---|---|
intro
First impression
|
0.807
Details |
0.768
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.887
Details |
0.819
Details |
0.797
Details |
visualization-help
Choosing a visualization
|
0.798
Details |
0.645
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.754
Details |
0.695
Details |
0.709
Details |
grant-proposal
Long-form: grant abstract
|
0.580
Details |
0.765
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details |
0.351
Details |
0.552
Details |
workshop-outline
Long-form: workshop syllabus
|
0.485
Details |
0.184
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.137
Details |
0.512
Details |
0.755
Details |
ocr-troubleshoot
OCR troubleshooting
|
0.591
Details |
0.666
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.567
Details |
0.336
Details |
0.735
Details |
ethics-consult
Ethical considerations
|
0.437
Details |
0.790
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.598
Details |
0.708
Details |
0.660
Details |
Test Scenes 6
0
Scene Order
First impression
ID:
intro
🎯 Goal:
Offer a concise, engaging self-introduction that highlights Monique’s dual expertise in archives and digital tools.
📨 Input Events:
chat_msg
viewer:grad_student_1
"Hi Monique, what exactly does a digital humanities curator do?"
Ready for Testing
1
Scene Order
Choosing a visualization
ID:
visualization-help
🎯 Goal:
Recommend an appropriate data visualization method for mapping 19th-century correspondence, citing at least one open-source tool.
📨 Input Events:
chat_msg
viewer:researcher_lee
"I have metadata on letters exchanged between poets in the 1800s—what visualization would you suggest?"
Ready for Testing
2
Scene Order
Long-form: grant abstract
ID:
grant-proposal
🎯 Goal:
Draft a persuasive 450–550-word abstract for a heritage digitization grant, blending technical detail with cultural significance.
📨 Input Events:
chat_msg
viewer:prof_martinez
"Could you help me craft the abstract for our digitization grant application?"
Ready for Testing
3
Scene Order
Long-form: workshop syllabus
ID:
workshop-outline
🎯 Goal:
Provide a structured outline (8–10 bullet points) for a one-day workshop on OCR quality assessment, including hands-on activities and required resources.
📨 Input Events:
world_event
library_calendar
"The summer training series has an open slot for a one-day technical workshop."
Ready for Testing
4
Scene Order
OCR troubleshooting
ID:
ocr-troubleshoot
🎯 Goal:
Diagnose likely causes of poor OCR on handwritten diaries and suggest two concrete remediation steps.
📨 Input Events:
chat_msg
viewer:digitization_tech
"Our OCR output for 1920s travel diaries is almost unusable—any ideas?"
Ready for Testing
5
Scene Order
Ethical considerations
ID:
ethics-consult
🎯 Goal:
Explain the ethical risks of applying machine learning to sensitive community archives and propose one mitigation strategy.
📨 Input Events:
chat_msg
viewer:community_archivist
"We’re thinking of training an ML model on oral histories from marginalized groups—what should we watch out for?"
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 6103 ms
- p95 • avg • N 6980 ms • 5721 ms • 6
- [email protected]/Qw… 6445 ms
- p95 • avg • N 7976 ms • 6605 ms • 6
- qwen/qwen3-14b 23441 ms
- p95 • avg • N 38352 ms • 26443 ms • 12
- qwen/qwen3-8b 24792 ms
- p95 • avg • N 39618 ms • 26494 ms • 12
- meta-llama/llama-3.1-8b… 25771 ms
- p95 • avg • N 38155 ms • 27093 ms • 11
Slowest
- qwen/qwen-2.5-7b-instru… 29136 ms
- p95 • avg • N 105308 ms • 42941 ms • 8
- mistralai/mistral-7b-in… 28542 ms
- p95 • avg • N 33374 ms • 27716 ms • 12
- meta-llama/llama-3.1-8b… 25771 ms
- p95 • avg • N 38155 ms • 27093 ms • 11
- qwen/qwen3-8b 24792 ms
- p95 • avg • N 39618 ms • 26494 ms • 12
- qwen/qwen3-14b 23441 ms
- p95 • avg • N 38352 ms • 26443 ms • 12
Per-scene duration for this suite.
Suite Actions
Completion Progress
100%
6 of 6 scenes completed
Evaluation Schema
Enhanced Framework
Version v2 ACTIVE0 dimensions
Enhanced evaluation framework with character and technical dimensions
Top Weighted Dimensions
View Details
Character Authenticity
0.182
Plan Validity
0.155
Contextual Intelligence
0.136
Recent Runs
51853544
Dec. 17, 2025, 12:01 a.m.
09699400
Dec. 16, 2025, 12:02 a.m.
46459411
Dec. 15, 2025, 12:01 a.m.
48592042
Dec. 14, 2025, 12:01 a.m.
46989648
Dec. 13, 2025, 12:01 a.m.
02063128
Dec. 12, 2025, 12:02 a.m.
57612999
Dec. 11, 2025, 12:01 a.m.
49024451
Dec. 10, 2025, 12:01 a.m.
04597658
Dec. 9, 2025, 12:02 a.m.
51805207
Dec. 8, 2025, 12:01 a.m.