List your company

Deductive Reasoning on BBH Ded.

92.5AccuracyBranching
66.18873.01979.8586.681Apr 22, 2026

Evaluation results

MethodLinks
BranchingModel=Phi-4-reasoning-plus (13B)2026.0492.5
Beam Search (BL)Model=Phi-4-reasoning-plus (13B)2026.0491.3
Beam Search (BL)Model=Gemma-3-12b-it2026.0489.2
SteeringModel=Phi-4-reasoning-plus (13B)2026.0488.7
SteeringModel=Gemma-3-12b-it2026.0487.8
Greedy (BL)Model=Gemma-3-12b-it2026.0487.5
BranchingModel=Gemma-3-12b-it2026.0487.4
TTPOModel=Gemma-3-12b-it2026.0486.4
TTPOModel=Phi-4-reasoning-plus (13B)2026.0486.4
Beam SearchModel=Gemma-3-4b-it, beam size=52026.0477
Beam Search (BL)Model=Gemma-3-4b-it2026.0477
TTPOModel=Gemma-3-4b-it2026.0475.9
TTPOModel=Gemma-3-4b-it2026.0475.9
Self-ConsistencyModel=Gemma-3-4b-it, n=52026.0475.5
SteeringModel=Gemma-3-4b-it2026.0475.5
GreedyModel=Gemma-3-4b-it2026.0475.3
Greedy (BL)Model=Gemma-3-4b-it2026.0475.3
BranchingModel=Gemma-3-4b-it2026.0474.7
Beam SearchModel=Phi-4-mini-instruct (4B), beam size=52026.0473.8
Beam Search (BL)Model=Phi-4-mini-instruct (4B)2026.0473.8
Self-ConsistencyModel=Phi-4-mini-instruct (4B), n=52026.0473.6
Greedy (BL)Model=Phi-4-reasoning-plus (13B)2026.0470.2
BranchingModel=Phi-4-mini-instruct (4B)2026.0469.1
BranchingModel=Phi-4-mini-instruct (4B)2026.0469.1
SteeringModel=Phi-4-mini-instruct (4B)2026.0468.8
GreedyModel=Phi-4-mini-instruct (4B)2026.0467.3
Greedy (BL)Model=Phi-4-mini-instruct (4B)2026.0467.3
TTPOModel=Phi-4-mini-instruct (4B)2026.0467.2