List your company

Group-level distractor generation on Discrete Math 40

36.43RecallDeepseek-V3
19.342823.778928.21532.6511Aug 15, 2025

Evaluation results

MethodLinks
Deepseek-V3Framework=MCTS-guided reasoning reconstruction, Backbone=Deepseek-V32025.0836.43
Claude-4-SonnetFramework=MCTS-guided reasoning reconstruction, Backbone=Claude-4-Sonnet2025.0834.29
GPT-3.5-turboFramework=MCTS-guided reasoning reconstruction, Backbone=GPT-3.5-turbo2025.0830.71
GPT-4oFramework=MCTS-guided reasoning reconstruction, Backbone=GPT-4o2025.0829.29
D-GEN2025.0828.57
LLaMA-3-70BFramework=MCTS-guided reasoning reconstruction, Backbone=LLaMA-3-70B2025.0828.57
DiVERT2025.0820.71
O&A2025.0820