List your company
ResearchFormerly WizWand

Accuracy on ARC-C (Question Answering)

86Accuracy (ARC-C)ExpGraph
17.131235.010652.8970.7694Oct 29, 2025Dec 8, 2025Jan 17, 2026Feb 26, 2026Apr 7, 2026May 17, 2026Jun 27, 2026
Updated 4mo ago

Evaluation results

MethodLinks
ExpGraphBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=LLM-Centric Experience Learning Baselines2026.0586
MemRLBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0584.34
Mem0Backbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0583.33
IRCoTBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=LLM-Centric Experience Learning Baselines2026.0583.08
ExpeLBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0582.58
ReasoningBankBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0582.32
S3Backbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=LLM-Centric Experience Learning Baselines2026.0582
LightMemBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0581.57
AWMBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0581.57
Search-o1Backbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=LLM-Centric Experience Learning Baselines2026.0581.31
ExpGraphBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=LLM-Centric Experience Learning Baselines2026.0574
AWMBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0572.47
ReasoningBankBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0571.72
LightMemBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0571.46
No MemoryBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=No Memory2026.0570.2
Mem0Backbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0567.93
MemRLBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0567.42
Search-o1Backbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=LLM-Centric Experience Learning Baselines2026.0566.67
S3Backbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=LLM-Centric Experience Learning Baselines2026.0565
Qwen3Model Version=4B, Architecture=Dense, # Params=4.0B, # Tokens=36T2025.1063.65
DreamReasoner-8BType=Block Diffusion2026.0663.5
IRCoTBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=LLM-Centric Experience Learning Baselines2026.0562.63
Gemma3Model Version=4B, Architecture=Dense, # Params=4.0B, # Tokens=4T2025.1060.92
OuroModel Version=1.4B R4, Architecture=LoopLM, # Params=1.4B, # Tokens=7.7T, recurrent steps=42025.1060.92
Dream-v0-7BType=Diffusion2026.0659.8
Qwen3-8BType=AR2026.0656.9
Qwen3Model Version=1.7B, Architecture=Dense, # Params=1.7B, # Tokens=36T2025.1055.72
Qwen2.5Model Version=3B, Architecture=Dense, # Params=3.0B, # Tokens=18T2025.1055.46
Qwen2.5Model Version=1.5B, Architecture=Dense, # Params=1.5B, # Tokens=18T2025.1054.44
OLMo3-7BType=AR2026.0653.5
Llama3.2Model Version=3B, Architecture=Dense, # Params=3.0B, # Tokens=9T2025.1052.47
MiMo-7BType=AR2026.0651.6
No MemoryBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=No Memory2026.0551.52
Qwen2.5-7BType=AR2026.0651.5
LLaDA-8BType=Diffusion2026.0647.5
ExpeLBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=Retrieval-Centric Experience Learning Baselines2026.0544.7
Llama3.2Model Version=1.2B, Architecture=Dense, # Params=1.0B, # Tokens=9T2025.1041.98
Gemma3Model Version=1B, Architecture=Dense, # Params=1.0B, # Tokens=2T2025.1039.25
TaDABackbone=Llama-2-7b, Number of samples=5002026.0637.4
LinearBackbone=Llama-2-7b, Number of samples=5002026.0636
Mag. PruningBackbone=Llama-2-7b, Number of samples=5002026.0635.8
TIESBackbone=Llama-2-7b, Number of samples=5002026.0634.4
Domain onlyBackbone=Llama-2-7b, Number of samples=5002026.0633.6
DARE-TIESBackbone=Llama-2-7b, Number of samples=500, Evaluation protocol=mean over 3 random seeds2026.0633.6
DeltaNetevaluation_mode=zero-shot, parameters=2.8B, context_length=< 2K tokens2025.1132.85
Gated DeltaNetevaluation_mode=zero-shot, parameters=2.8B, context_length=< 2K tokens2025.1132.59
Gated KalmaNetevaluation_mode=zero-shot, parameters=2.8B, context_length=< 2K tokens2025.1132.51
Transformerevaluation_mode=zero-shot, parameters=2.8B, context_length=< 2K tokens2025.1132.25
Mamba2evaluation_mode=zero-shot, parameters=2.8B, context_length=< 2K tokens2025.1132.24
Task onlyBackbone=Llama-2-7b, Number of samples=5002026.0631.6
Base modelBackbone=Llama-2-7b, Number of samples=5002026.0631
SVD MergeBackbone=Llama-2-7b, Number of samples=5002026.0628.4
Gated Linear Attentionevaluation_mode=zero-shot, parameters=2.8B, context_length=< 2K tokens2025.1127.82
DCDM (MoE)Scale=1.5B, Active Params=1.2B (2.8B), In-context examples=0-shot2026.0527.5
DARE-LinBackbone=Llama-2-7b, Number of samples=500, Evaluation protocol=mean over 3 random seeds2026.0627.2
DCDMScale=1.5B, Active Params=1.5B, In-context examples=0-shot2026.0527.13
MDLMScale=1.5B, Active Params=1.5B, In-context examples=0-shot2026.0526.79
BDLMScale=1.5B, Active Params=1.5B, In-context examples=0-shot2026.0526.71
Task Arith.Backbone=Llama-2-7b, Number of samples=5002026.0626.4
DCDM (MoE)Scale=0.5B, Active Params=0.4B (0.8B), In-context examples=0-shot2026.0524.74
DCDMScale=0.5B, Active Params=0.5B, In-context examples=0-shot2026.0524.49
MDLMScale=0.5B, Active Params=0.5B, In-context examples=0-shot2026.0523.29
BDLMScale=0.5B, Active Params=0.5B, In-context examples=0-shot2026.0523.21
DLR +CoLAModel Scale=1B, Rank (r)=512, Evaluation Protocol=Zero-shot2026.0622.29
Full-RankModel Scale=1B, Rank (r)=512, Evaluation Protocol=Zero-shot2026.0622.06
DLR +Low-RankModel Scale=1B, Rank (r)=512, Evaluation Protocol=Zero-shot2026.0621.59
Low-RankModel Scale=1B, Rank (r)=512, Evaluation Protocol=Zero-shot2026.0619.78