Financial Analysis on Financial Analysis Benchmark
81FiNER AccuracyACE
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ACEBase LLM=GPT-5.1, Adaptation Protocol=Offline Adaptation, GT Labels=✓2025.10 | 81 | 84.5 | 82.8 | |
| ACEBase LLM=DeepSeek-V3.1-671B, Adaptation Setting=Offline, GT Labels=true2025.10 | 78.3 | 85.5 | 81.9 | |
| ACEBase LLM=GPT-5.1, Adaptation Protocol=Offline Adaptation, GT Labels=✗2025.10 | 78.2 | 76.5 | 77.4 | |
| ACEBase LLM=DeepSeek-V3.1-671B, Adaptation Setting=Online, GT Labels=true2025.10 | 76.7 | 76.5 | 76.6 | |
| ACEBase LLM=GPT-5.1, Adaptation Protocol=Online Adaptation, GT Labels=✗2025.10 | 75.2 | 79 | 77.1 | |
| GEPABase LLM=GPT-5.1, Adaptation Protocol=Offline Adaptation, GT Labels=✓2025.10 | 74.5 | 73 | 73.8 | |
| DC (CU)Base LLM=DeepSeek-V3.1-671B, Adaptation Setting=Online, GT Labels=true2025.10 | 74.2 | 69.5 | 71.8 | |
| ACEBase LLM=GPT-OSS-120B, Adaptation Type=Offline Adaptation, GT Labels=✓2025.10 | 73.8 | 88.5 | 81.2 | |
| GEPABase LLM=DeepSeek-V3.1-671B, Adaptation Setting=Offline, GT Labels=true2025.10 | 73.5 | 71.5 | 72.5 | |
| Base LLMBase LLM=GPT-5.12025.10 | 73.5 | 73 | 73.3 | |
| MIPROv2Base LLM=DeepSeek-V3.1-671B, Adaptation Setting=Offline, GT Labels=true2025.10 | 72.4 | 69.5 | 70.9 | |
| ICLBase LLM=DeepSeek-V3.1-671B, Adaptation Setting=Offline, GT Labels=true2025.10 | 72.3 | 67 | 69.6 | |
| ACEBase LLM=DeepSeek-V3.1-671B, Adaptation Setting=Offline, GT Labels=false2025.10 | 71.1 | 83 | 77.1 | |
| Base LLMBase LLM=DeepSeek-V3.1-671B2025.10 | 70.7 | 67.5 | 69.1 | |
| ACEBase LLM=GPT-OSS-120B, Adaptation Type=Online Adaptation, GT Labels=✗2025.10 | 70.5 | 85 | 77.8 | |
| DCBase LLM=GPT-5.1, Adaptation Protocol=Online Adaptation, GT Labels=✗2025.10 | 70 | 69 | 69.5 | |
| ACEBase LLM=GPT-OSS-120B, Adaptation Type=Offline Adaptation, GT Labels=✗2025.10 | 69.7 | 84 | 76.9 | |
| DC (CU)Base LLM=DeepSeek-V3.1-671B, Adaptation Setting=Online, GT Labels=false2025.10 | 68.3 | 62.5 | 65.4 | |
| GEPABase LLM=GPT-OSS-120B, Adaptation Type=Offline Adaptation, GT Labels=✓2025.10 | 67.9 | 71.5 | 69.7 | |
| ACEBase LLM=DeepSeek-V3.1-671B, Adaptation Setting=Online, GT Labels=false2025.10 | 67.3 | 78.5 | 72.9 | |
| Base LLMBase LLM=GPT-OSS-120B2025.10 | 66.6 | 71.5 | 69.1 | |
| DCBase LLM=GPT-OSS-120B, Adaptation Type=Online Adaptation, GT Labels=✗2025.10 | 55.8 | 66 | 60.9 |