List your company

Logical Reasoning on ZebraLogic (test)

92.2Grid AccuracyIn-place
-3.68821.20646.170.994Oct 1, 2025

Evaluation results

MethodLinks
In-placeBackbone=Qwen, Number of turns=9, Feedback=top-42025.1092.2———
In-placeBackbone=Gemma, Number of turns=9, Feedback=top-42025.1090.6———
In-placeBackbone=Qwen, Number of turns=8, Feedback=top-42025.1089.5———
In-placeBackbone=Gemma, Number of turns=8, Feedback=top-42025.1086.8———
In-placeBackbone=Qwen, Number of turns=7, Feedback=top-42025.1085———
In-placeBackbone=Gemma, Number of turns=7, Feedback=top-42025.1081.6———
In-placeBackbone=Llama, Number of turns=9, Feedback=top-42025.1079———
In-placeBackbone=Qwen, Number of turns=6, Feedback=top-42025.1077.9———
Multi-turnBackbone=Gemma, Number of turns=9, Feedback=top-42025.1077.1———
In-placeBackbone=Llama, Number of turns=8, Feedback=top-42025.1073.8———
Multi-turnBackbone=Gemma, Number of turns=8, Feedback=top-42025.1072.8———
In-placeBackbone=Gemma, Number of turns=6, Feedback=top-42025.1071.2———
Multi-turnBackbone=Qwen, Number of turns=9, Feedback=top-42025.1069———
MemoryBackbone=Llama, Number of turns=9, Feedback=top-42025.1068.2———
In-placeBackbone=Llama, Number of turns=7, Feedback=top-42025.1067.3———
In-placeBackbone=Qwen, Number of turns=5, Feedback=top-42025.1067.2———
MemoryBackbone=Qwen, Number of turns=9, Feedback=top-42025.1066.5———
Multi-turnBackbone=Llama, Number of turns=9, Feedback=top-42025.1065.2———
Multi-turnBackbone=Gemma, Number of turns=7, Feedback=top-42025.1065———
Multi-turnBackbone=Qwen, Number of turns=8, Feedback=top-42025.1064.5———
MemoryBackbone=Llama, Number of turns=8, Feedback=top-42025.1062.5———
Multi-turnBackbone=Llama, Number of turns=8, Feedback=top-42025.1061.2———
MemoryBackbone=Qwen, Number of turns=8, Feedback=top-42025.1060———
In-placeBackbone=Llama, Number of turns=6, Feedback=top-42025.1059———
In-placeBackbone=Gemma, Number of turns=5, Feedback=top-42025.1058.2———
Multi-turnBackbone=Qwen, Number of turns=7, Feedback=top-42025.1056.6———
Multi-turnBackbone=Gemma, Number of turns=6, Feedback=top-42025.1055.6———
Multi-turnBackbone=Llama, Number of turns=7, Feedback=top-42025.1055.1———
MemoryBackbone=Llama, Number of turns=7, Feedback=top-42025.1054.3———
In-placeBackbone=Qwen, Number of turns=4, Feedback=top-42025.1053.4———
MemoryBackbone=Qwen, Number of turns=7, Feedback=top-42025.1052———
In-placeBackbone=Llama, Number of turns=5, Feedback=top-42025.1047.8———
Multi-turnBackbone=Llama, Number of turns=6, Feedback=top-42025.1047———
Multi-turnBackbone=Qwen, Number of turns=6, Feedback=top-42025.1046.8———
MemoryBackbone=Llama, Number of turns=6, Feedback=top-42025.1045.7———
In-placeBackbone=Gemma, Number of turns=4, Feedback=top-42025.1044.1———
MemoryBackbone=Qwen, Number of turns=6, Feedback=top-42025.1043.4———
Multi-turnBackbone=Gemma, Number of turns=5, Feedback=top-42025.1042.1———
MemoryBackbone=Llama, Number of turns=5, Feedback=top-42025.1036.9———
In-placeBackbone=Llama, Number of turns=4, Feedback=top-42025.1035.9———
Multi-turnBackbone=Llama, Number of turns=5, Feedback=top-42025.1035.8———
Multi-turnBackbone=Qwen, Number of turns=5, Feedback=top-42025.1035.6———
MemoryBackbone=Qwen, Number of turns=5, Feedback=top-42025.1034.8———
In-placeBackbone=Qwen, Number of turns=3, Feedback=top-42025.1034.1———
Multi-turnBackbone=Gemma, Number of turns=4, Feedback=top-42025.1027.1———
In-placeBackbone=Gemma, Number of turns=3, Feedback=top-42025.1026.5———
MemoryBackbone=Llama, Number of turns=4, Feedback=top-42025.1026.4———
MemoryBackbone=Qwen, Number of turns=4, Feedback=top-42025.1025.1———
Multi-turnBackbone=Qwen, Number of turns=4, Feedback=top-42025.1024.9———
MemoryBackbone=Gemma, Number of turns=9, Feedback=top-42025.1023.4———
Multi-turnBackbone=Llama, Number of turns=4, Feedback=top-42025.1022.5———
In-placeBackbone=Llama, Number of turns=3, Feedback=top-42025.1022.3———
MemoryBackbone=Gemma, Number of turns=8, Feedback=top-42025.1019.9———
MemoryBackbone=Gemma, Number of turns=7, Feedback=top-42025.1016.7———
MemoryBackbone=Qwen, Number of turns=3, Feedback=top-42025.1016.6———
MemoryBackbone=Llama, Number of turns=3, Feedback=top-42025.1016———
Multi-turnBackbone=Qwen, Number of turns=3, Feedback=top-42025.1015———
Multi-turnBackbone=Gemma, Number of turns=3, Feedback=top-42025.1014.8———
In-placeBackbone=Qwen, Number of turns=2, Feedback=top-42025.1014.4———
MemoryBackbone=Gemma, Number of turns=6, Feedback=top-42025.1013.5———
Multi-turnBackbone=Llama, Number of turns=3, Feedback=top-42025.1012.6———
In-placeBackbone=Gemma, Number of turns=2, Feedback=top-42025.1011.5———
MemoryBackbone=Gemma, Number of turns=5, Feedback=top-42025.1010.5———
In-placeBackbone=Llama, Number of turns=2, Feedback=top-42025.1010.5———
Multi-turnBackbone=Gemma, Number of turns=2, Feedback=top-42025.107.8———
MemoryBackbone=Gemma, Number of turns=4, Feedback=top-42025.107.8———
MemoryBackbone=Llama, Number of turns=2, Feedback=top-42025.107.6———
MemoryBackbone=Qwen, Number of turns=2, Feedback=top-42025.107.3———
Multi-turnBackbone=Qwen, Number of turns=2, Feedback=top-42025.107.1———
Multi-turnBackbone=Llama, Number of turns=2, Feedback=top-42025.105.6———
MemoryBackbone=Gemma, Number of turns=3, Feedback=top-42025.105.2———
In-placeBackbone=Qwen, Number of turns=1, Feedback=top-42025.103.6———
In-placeBackbone=Llama, Number of turns=1, Feedback=top-42025.103.4———
In-placeBackbone=Gemma, Number of turns=1, Feedback=top-42025.103.1———
Multi-turnBackbone=Gemma, Number of turns=1, Feedback=top-42025.102.8———
MemoryBackbone=Gemma, Number of turns=2, Feedback=top-42025.102.8———
MemoryBackbone=Qwen, Number of turns=1, Feedback=top-42025.102.5———
MemoryBackbone=Llama, Number of turns=1, Feedback=top-42025.102.3———
Multi-turnBackbone=Qwen, Number of turns=1, Feedback=top-42025.102.2———
Multi-turnBackbone=Llama, Number of turns=1, Feedback=top-42025.101.7———
MemoryBackbone=Gemma, Number of turns=1, Feedback=top-42025.100.7———
Multi-turnBackbone=Llama, Number of turns=0, Feedback=top-42025.100.6———
MemoryBackbone=Llama, Number of turns=0, Feedback=top-42025.100.6———
In-placeBackbone=Llama, Number of turns=0, Feedback=top-42025.100.6———
MemoryBackbone=Qwen, Number of turns=0, Feedback=top-42025.100.4———
In-placeBackbone=Qwen, Number of turns=0, Feedback=top-42025.100.4———
Multi-turnBackbone=Qwen, Number of turns=0, Feedback=top-42025.100.3———
Multi-turnBackbone=Gemma, Number of turns=0, Feedback=top-42025.100———
MemoryBackbone=Gemma, Number of turns=0, Feedback=top-42025.100———
In-placeBackbone=Gemma, Number of turns=0, Feedback=top-42025.100———
G-PACScore type=Logits-based score, LLM Backbone=Llama-based, Alpha=0.05, Epsilon=0.22026.01—11.43013.23
G-PACScore type=Verbalized score, LLM Backbone=Llama-based, Alpha=0.05, Epsilon=0.22026.01—11.130-9.52
G-PACScore type=Router-based score, LLM Backbone=Llama-based, Alpha=0.05, Epsilon=0.22026.01—11.13025.98
PACScore type=Logits-based score, LLM Backbone=Llama-based, Alpha=0.05, Epsilon=0.22026.01—17.3316.223.72
PACScore type=Verbalized score, LLM Backbone=Llama-based, Alpha=0.05, Epsilon=0.22026.01—17.0515.37-30.68
PACScore type=Router-based score, LLM Backbone=Llama-based, Alpha=0.05, Epsilon=0.22026.01—17.215.5525.22