List your company

Murder Mystery Reasoning on MUSR murder (test)

75AccuracyInduced
43.851.96068.1Jun 2, 2026

Evaluation results

MethodLinks
InducedLLM Backbone=DeepSeek-V32026.0675
Induced2026.0675
Hand-designedLLM Backbone=DeepSeek-V32026.0669
Hand2026.0669
Program-of-ThoughtsLLM Backbone=DeepSeek-V32026.0667
Program-of-Thoughts2026.0667
Chain-of-ThoughtLLM Backbone=DeepSeek-V32026.0660
CoT2026.0660
Agent Workflow MemoryLLM Backbone=DeepSeek-V32026.0647
AWM2026.0647
ReActLLM Backbone=DeepSeek-V32026.0645
ReAct2026.0645