Long-form Dialogue on LoCoMo
37.24EMDYCP
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| DYCPAgent=Claude 4.02026.01 | 37.24 | 7.48 | 3.41 | 6.83 | 90.45 | 1.85 | |
| DYCPAgent=GPT-4.12026.01 | 35.46 | 12.45 | 5.91 | 11.56 | 91.46 | 0.97 | |
| Full ContextAgent=Claude 4.02026.01 | 34.69 | 6.89 | 2.98 | 6.24 | 89.57 | 5.3 | |
| Full ContextAgent=GPT-4.12026.01 | 33.67 | 12.31 | 5.63 | 11.35 | 92.06 | 2.41 | |
| DYCPAgent=Claude 3.72026.01 | 31.89 | 7.98 | 3.16 | 7.08 | 83.05 | 1.52 | |
| DYCPAgent=GPT-4o mini2026.01 | 28.32 | 16.12 | 6.69 | 14.79 | 80.44 | 0.88 | |
| DYCPAgent=GPT-4o2026.01 | 26.53 | 14.72 | 6.22 | 13.51 | 83.27 | 1.1 | |
| Full ContextAgent=Claude 3.72026.01 | 26.28 | 6.69 | 2.64 | 5.99 | 74.21 | 5.54 | |
| DYCPAgent=Mistral-Nemo-Instruct (128k)2026.01 | 25.26 | 23.44 | 10.39 | 21.84 | 79.04 | 1.52 | |
| DYCPAgent=Qwen2.5-7B-Instruct (1M)2026.01 | 25 | 9.63 | 4.15 | 8.88 | 75.96 | 0.98 | |
| Full ContextAgent=GPT-4o mini2026.01 | 24.49 | 15.04 | 6.01 | 13.91 | 74.96 | 2.43 | |
| Full ContextAgent=GPT-4o2026.01 | 24.23 | 12.28 | 5.16 | 11.28 | 75.13 | 2.32 | |
| Full ContextAgent=Qwen2.5-7B-Instruct (1M)2026.01 | 22.96 | 9.02 | 3.9 | 8.3 | 72.57 | 6.56 | |
| SCM4LLMsAgent=Claude 3.72026.01 | 22.19 | 5.8 | 1.86 | 5.25 | 62.82 | 6.65 | |
| Full ContextAgent=Mistral-Nemo-Instruct (128k)2026.01 | 21.17 | 20.59 | 8.81 | 19.33 | 66.75 | 10.18 | |
| SeComAgent=GPT-4o2026.01 | 21.17 | 12.45 | 4.7 | 11.32 | 67.19 | 1.26 | |
| SeComAgent=Claude 3.72026.01 | 21.17 | 6.04 | 2.11 | 5.38 | 60.88 | 1.53 | |
| SeComAgent=GPT-4o mini2026.01 | 19.64 | 12.01 | 4.5 | 10.97 | 66.09 | 0.88 | |
| SCM4LLMsAgent=GPT-4o mini2026.01 | 19.13 | 13.64 | 5.4 | 12.59 | 66.47 | 2.1 | |
| CondMemAgent=Qwen2.5-7B-Instruct (1M)2026.01 | 16.84 | 12.39 | 4.83 | 11.49 | 57.07 | 7.45 | |
| CondMemAgent=Mistral-Nemo-Instruct (128k)2026.01 | 16.58 | 16.53 | 5.96 | 15.19 | 56.36 | 3.25 | |
| SCM4LLMsAgent=GPT-4o2026.01 | 12.5 | 9.84 | 3.66 | 9.03 | 49.6 | 2.01 | |
| MemoChatAgent=GPT-4o mini2026.01 | 9.69 | 7.69 | 2.65 | 6.83 | 45.62 | 2.46 | |
| MemoChatAgent=GPT-4o2026.01 | 6.89 | 7.05 | 1.79 | 6.19 | 43.39 | 2.24 | |
| No ContextAgent=Mistral-Nemo-Instruct (128k)2026.01 | 4.08 | 5.24 | 0.92 | 4.64 | 23.9 | 0.05 | |
| MemoChatAgent=Claude 3.72026.01 | 3.83 | 3.58 | 0.58 | 3.08 | 30.15 | 5.53 | |
| No ContextAgent=GPT-4o2026.01 | 1.79 | 2.73 | 0.33 | 2.36 | 15.57 | 0.48 | |
| No ContextAgent=GPT-4o mini2026.01 | 1.79 | 3.09 | 0.33 | 2.66 | 17.19 | 0.48 | |
| No ContextAgent=GPT-4.12026.01 | 1.79 | 4.76 | 0.66 | 3.95 | 25.21 | 0.43 | |
| No ContextAgent=Qwen2.5-7B-Instruct (1M)2026.01 | 1.28 | 3.23 | 0.31 | 2.8 | 21.49 | 0.04 | |
| No ContextAgent=Claude 4.02026.01 | 0.77 | 2.83 | 0.21 | 2.46 | 13.3 | 1.15 | |
| No ContextAgent=Claude 3.72026.01 | 0.51 | 2.47 | 0.2 | 2.18 | 13.3 | 1.15 |