List your company

Social Interaction Simulation on EduMirror social interaction steps (144 steps) (test)

4.958NaturalnessEduMirror
3.3983.8034.2084.613Jun 6, 2026

Evaluation results

MethodLinks
EduMirrorLLM Backbone=GPT-4.12026.064.9584.958
EduMirrorLLM Backbone=Average2026.064.8024.821
EduMirrorLLM Backbone=Qwen32026.064.7924.824
EduMirrorLLM Backbone=DeepSeek2026.064.754.792
EduMirrorLLM Backbone=Gemini2026.064.7084.708
ReActLLM Backbone=GPT-4.12026.064.6674.86
LLMobLLM Backbone=GPT-4.12026.064.6254.875
LLMobLLM Backbone=Average2026.064.2294.469
ReActLLM Backbone=Gemini2026.064.2084.417
ReActLLM Backbone=Average2026.064.2084.496
LLMobLLM Backbone=Gemini2026.064.1674.292
LLMobLLM Backbone=DeepSeek2026.064.0834.375
JAG-ConcordiaLLM Backbone=Qwen32026.064.0834.192
D2ALLM Backbone=Gemini2026.064.0424.181
LLMobLLM Backbone=Qwen32026.064.0424.333
ReActLLM Backbone=DeepSeek2026.0644.5
JAG-ConcordiaLLM Backbone=GPT-4.12026.063.9584.133
ReActLLM Backbone=Qwen32026.063.9584.208
BabyAGILLM Backbone=Qwen32026.063.9583.958
D2ALLM Backbone=DeepSeek2026.063.9254.1
JAG-ConcordiaLLM Backbone=Average2026.063.9224.063
JAG-ConcordiaLLM Backbone=Gemini2026.063.8854.052
D2ALLM Backbone=Average2026.063.8844.083
BabyAGILLM Backbone=DeepSeek2026.063.8754.042
D2ALLM Backbone=Qwen32026.063.8674.117
JAG-ConcordiaLLM Backbone=DeepSeek2026.063.763.875
D2ALLM Backbone=GPT-4.12026.063.73.933
BabyAGILLM Backbone=Average2026.063.6983.875
BabyAGILLM Backbone=Gemini2026.063.53.708
BabyAGILLM Backbone=GPT-4.12026.063.4583.792