Datasets·AI·aita_human_vs_gemini_3lens

AITA human vs Gemini

AITA: Human vs Gemini . PsyProxy quantitative evaluation. The excerpts reflect a range of interpersonal conflicts and moral dilemmas, often framed in the context of relationships, personal boundaries, and social etiquette. A recurring theme is the concept of **accountability**, as individuals grapple with the consequences of their actions and the impact on others, particularly in familial and romantic settings. The tone varies from confrontational to reflective, with many texts urging self-awareness and consideration for others' feelings. Issues such as **trust**, **communication**, and **expectations** frequently arise, highlighting the complexities of navigating social dynamics and the importance of empathy. Additionally, the discussions often touch on societal norms and personal autonomy, particularly in relation to **friendship** and **romantic relationships**, emphasizing the delicate balance between honesty and kindness. [Summary on 50 random texts by ChatGPT 4o Mini].

Distribution of Human vs Gemini AITA-style writing
1
2
1,817 at floor1,919 at ceiling
3,736
items
747
holdout n
Human vs Gemini AITA-style writing
target
Binary
kind
15
systems compared
Criterion validity

Reported holdout systems from the verified card

Binary classification uses FVE as the task-primary metric. Secondary columns keep the companion metrics visible so binary, ordinal, regression, and multiclass cards are not compared through one flattened score.

Model-family mix
PsyProxy · 4Baseline · 11
SystemFamilyVariantFVEAUCF1Primary scale
psyproxyPsyProxy — Health Lens v0.9 · 1100d
PsyProxypermissive0.9200.9980.978
psyproxyPsyProxy — Technology Lens v0.5 · 800d
PsyProxypermissive0.9080.9980.975
psyproxyPsyProxy — Behavioral Sciences Lens v0.5 · 1000d
PsyProxypermissive0.9020.9970.974
psyproxyPsyProxy — Sociology Lens v0.5 · 1000d
PsyProxypermissive0.0040.9810.980
baselineIMDB movie reviews (ACL) binary · via best lens
Baselinepermissive
baselineDisneyland TripAdvisor reviews continuous · via best lens
Baselinepermissive
baselineDouban movie reviews (Chinese) ordinal · via best lens
Baselinepermissive
baselineLIAR fact-check statements ordinal · via best lens
Baselinepermissive
baselineDisneyland TripAdvisor reviews binary · via best lens
Baselinepermissive
baselineSentiment140 tweets binary · via best lens
Baselinepermissive
baselineAmazon Video-Games reviews continuous · via best lens
Baselinepermissive
baselineAmazon Video-Games reviews ordinal · via best lens
Baselinepermissive
baselineDruglib drug reviews regression · via best lens
Baselinepermissive
baselineAmazon Video-Games reviews binary · via best lens
Baselinepermissive
baselineDruglib drug reviews ordinal · via best lens
Baselinepermissive