List your company

Multi-turn Instruction Following on MT-Eval

93.62CSR+GraphIF
48.931260.533172.13583.7369Nov 13, 2025

Evaluation results

MethodLinks
+GraphIFBackbone Model=Llama-3.1-Storm-8B2025.1193.6281.393.3394.87
+GraphIFBackbone Model=Qwen2.5-7B-Instruct2025.1191.376.9691.0690.72
+GraphIFBackbone Model=Llama-3.1-8B-Instruct2025.1191.2780.3590.8792.58
+GraphIFBackbone Model=Llama-3.1-Tulu-3.1-8B2025.1188.4876.9688.0389.68
+GraphIFBackbone Model=Hermes-3-Llama-3.1-8B2025.1186.9670.4386.5286.64
LLM-onlyBackbone Model=Llama-3.1-Storm-8B2025.1180.3656.9681.3680.87
LLM-onlyBackbone Model=Qwen2.5-7B-Instruct2025.1180.2251.379.779.57
LLM-onlyBackbone Model=Hermes-3-Llama-3.1-8B2025.1176.7448.776.2176.09
LLM-onlyBackbone Model=Llama-3.1-Tulu-3.1-8B2025.1174.1344.3575.372.7
+MemoryBankBackbone Model=Llama-3.1-Tulu-3.1-8B2025.1173.743.0473.0373.45
+MemoryBankBackbone Model=Qwen2.5-7B-Instruct2025.1170.2930.8769.8467.88
+MemoryBankBackbone Model=Hermes-3-Llama-3.1-8B2025.1168.4134.3568.3369.07
+MemoryBankBackbone Model=Llama-3.1-8B-Instruct2025.1167.1731.7466.9767.74
LLM-onlyBackbone Model=Llama-3.1-8B-Instruct2025.1167.0327.5167.8864.22
+MemoryBankBackbone Model=Llama-3.1-Storm-8B2025.1166.1628.765.4566.35
+MemoChatBackbone Model=Qwen2.5-7B-Instruct2025.1159.4216.5258.7956.35
+MemoChatBackbone Model=Llama-3.1-8B-Instruct2025.1156.315.2255.4553.97
+MemoChatBackbone Model=Llama-3.1-Storm-8B2025.1155.3613.0454.3953.59
+MemoChatBackbone Model=Hermes-3-Llama-3.1-8B2025.1154.9311.353.9452.26
+MemoChatBackbone Model=Llama-3.1-Tulu-3.1-8B2025.1150.6519.5751.3649.83