List your company

Hallucination Detection on TriviaQA In-domain (test)

99.1AccuracyAURORA
55.62866.91478.289.486Jun 28, 2026

Evaluation results

MethodLinks
AURORALLM Backbone=Gemma3-12B-it2026.0699.1
HARPLLM Backbone=Ministral3-8B-Base2026.0699.05
HARPLLM Backbone=Qwen3-8B-it2026.0699
HARPLLM Backbone=Gemma3-12B-it2026.0698.65
AURORALLM Backbone=Ministral3-8B-Base2026.0698.45
AURORALLM Backbone=Llama-3-8B-it2026.0698.25
HARPLLM Backbone=Llama-3-8B-it2026.0698.2
AURORALLM Backbone=Qwen3-8B-it2026.0697.9
ICR probeLLM Backbone=Gemma3-12B-it2026.0697.65
ICR probeLLM Backbone=Qwen3-8B-it2026.0695
ICR probeLLM Backbone=Llama-3-8B-it2026.0693.75
ICR probeLLM Backbone=Ministral3-8B-Base2026.0693.4
InterrogateLLMLLM Backbone=Ministral3-8B-Base2026.0684.1
InterrogateLLMLLM Backbone=Qwen3-8B-it2026.0676.45
InterrogateLLMLLM Backbone=Gemma3-12B-it2026.0669.05
InterrogateLLMLLM Backbone=Llama-3-8B-it2026.0657.3