List your company

Memory Fidelity Evaluation on Memory-based Experiment Seen Features

-0.068P-DiffRDG
-0.090240.059880.210.36012Nov 28, 2025

Evaluation results

MethodLinks
RDGBackbone=GPT-4o-mini-2024-07-182025.11-0.06800.0454.3
RDGBackbone=deepseek-r12025.110.006-0.0030.042.1
RDGBackbone=GPT-4o-2024-08-062025.110.0100.0351.9
CDDBackbone=deepseek-r12025.110.023-0.0180.06545.2
RDGBackbone=qwen3-235b-a22b2025.110.028-0.0010.0351.1
CoTBackbone=deepseek-r12025.110.03-0.0530.0456.9
CDDBackbone=qwen3-235b-a22b2025.110.0350.0010.05825.3
CDDBackbone=GPT-4o-2024-08-062025.110.054-0.0020.04232.6
BaselineBackbone=deepseek-r12025.110.057-0.0410.083287.3
DDDBackbone=deepseek-r12025.110.068-0.0480.092312.6
BaselineBackbone=qwen3-235b-a22b2025.110.070.0020.074125.6
ExplicitBackbone=qwen3-235b-a22b2025.110.070.0020.057.8
ExplicitBackbone=GPT-4o-2024-08-062025.110.07200.0456.6
ExplicitBackbone=deepseek-r12025.110.0810.0030.0555.4
DDDBackbone=qwen3-235b-a22b2025.110.0820.0080.078142.8
CoTBackbone=GPT-4o-2024-08-062025.110.08800.0488.8
Self-debiasBackbone=deepseek-r12025.110.1-0.0080.0629.2
BaselineBackbone=GPT-4o-2024-08-062025.110.1200.05151
ExplicitBackbone=GPT-4o-mini-2024-07-182025.110.1230.0220.0489.8
Self-debiasBackbone=GPT-4o-2024-08-062025.110.1320.0790.05511
DDDBackbone=GPT-4o-2024-08-062025.110.1380.0120.055168.5
CDDBackbone=GPT-4o-mini-2024-07-182025.110.1450.0420.05585.7
CounterfactualBackbone=GPT-4o-2024-08-062025.110.1560.080.05812.5
CoTBackbone=GPT-4o-mini-2024-07-182025.110.1620.0450.05212.5
Self-debiasBackbone=GPT-4o-mini-2024-07-182025.110.1770.060.06515.3
Self-debiasBackbone=qwen3-235b-a22b2025.110.1770.0620.04918.8
CounterfactualBackbone=GPT-4o-mini-2024-07-182025.110.2170.0740.07218.7
CoTBackbone=qwen3-235b-a22b2025.110.2820.1580.05129.6
BaselineBackbone=GPT-4o-mini-2024-07-182025.110.3210.0960.065506.5
DDDBackbone=GPT-4o-mini-2024-07-182025.110.3580.1150.068558.4
CounterfactualBackbone=qwen3-235b-a22b2025.110.3970.2450.05131.5
CounterfactualBackbone=deepseek-r12025.110.4880.4990.05745.8