List your company
ResearchFormerly WizWand

inductive Prolog rule synthesis on SLR-Bench Medium tier 250 tasks 1

88.8AccuracyDF + CC
30.5645.6860.875.92May 6, 2026
Updated 6mo ago

Evaluation results

MethodLinks
DF + CCCategory=Hybrid (symbolic + LLM fallback)2026.0588.8
DF + CC + QOCategory=Hybrid (symbolic + LLM fallback)2026.0588.8
DF + QOCategory=Hybrid (symbolic + LLM fallback)2026.0586.4
DF-32Category=LLM-only methods2026.0584.4
o3†Category=LLM-only methods2026.0574
gpt-5†Category=LLM-only methods2026.0572
BoK + CCCategory=Hybrid (symbolic + LLM fallback)2026.0568.4
BoK + CC + QOCategory=Hybrid (symbolic + LLM fallback)2026.0568.4
BoK + QOCategory=Hybrid (symbolic + LLM fallback)2026.0562.8
BoKCategory=LLM-only methods2026.0557.6
CC SolverCategory=Symbolic solvers2026.0548.4
QO SolverCategory=Symbolic solvers2026.0534.4
Qwen3.6-35B-A3B (OpenHands)Category=LLM-only methods2026.0532.8