List your company

Financial Question Answering on FiQA

82.7AccuracyOurs
-3.30819.02141.3563.679Dec 10, 2025Dec 21, 2025Jan 1, 2026Jan 13, 2026Jan 24, 2026Feb 4, 2026Feb 16, 2026

Evaluation results

MethodLinks
Ours2025.1282.744.829.60.79
Agentlite2025.1277.539.525.20.7
Agent-safetybench2025.1276.338.1240.67
Aios2025.1274.636.823.10.65
Expel2025.1271.834.221.50.62
XTFBackbone=Deepseek-distilled-qwen, Parameters=14B, LoRA=true2026.0245.6———
TCBackbone=Deepseek-distilled-qwen, Parameters=14B, LoRA=true2026.0241.2———
SLMBackbone=Deepseek-distilled-qwen, Parameters=14B, LoRA=true2026.0240.7———
×2 EpBackbone=Deepseek-distilled-qwen, Parameters=14B, LoRA=true2026.0236.2———
XTFBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=false2026.0236.1———
NormalBackbone=Deepseek-distilled-qwen, Parameters=14B, LoRA=true2026.0235.8———
DABackbone=Deepseek-distilled-qwen, Parameters=14B, LoRA=true2026.0234.1———
TCBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=false2026.0231.9———
SLMBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=false2026.0231.6———
DFBackbone=Deepseek-distilled-qwen, Parameters=14B, LoRA=true2026.0231.5———
DABackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=false2026.0230.9———
XTFBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=true2026.0229.3———
×2 EpBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=false2026.0228.9———
NormalBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=false2026.0228.5———
DFBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=false2026.0228.3———
CABackbone=Deepseek-distilled-qwen, Parameters=14B, LoRA=true2026.0226.3———
XTFBackbone=Deepseek-distilled-qwen, Parameters=1.5B, LoRA=false2026.0225.9———
DABackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=true2026.0223.9———
SLMBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=true2026.0222.8———
×2 EpBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=true2026.0222.7———
NormalBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=true2026.0222.6———
TCBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=true2026.0222.1———
SLMBackbone=Deepseek-distilled-qwen, Parameters=1.5B, LoRA=false2026.0221.5———
TCBackbone=Deepseek-distilled-qwen, Parameters=1.5B, LoRA=false2026.0220.8———
DFBackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=true2026.0220.8———
DABackbone=Deepseek-distilled-qwen, Parameters=1.5B, LoRA=false2026.0219.2———
×2 EpBackbone=Deepseek-distilled-qwen, Parameters=1.5B, LoRA=false2026.0218.7———
XTFBackbone=Mistral, Parameters=7B, LoRA=false2026.0218.5———
NormalBackbone=Deepseek-distilled-qwen, Parameters=1.5B, LoRA=false2026.0218.4———
CABackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=false2026.0218.4———
CABackbone=Deepseek-distilled-qwen, Parameters=7B, LoRA=true2026.0218.4———
XTFBackbone=Mistral, Parameters=7B, LoRA=true2026.0217.2———
DFBackbone=Deepseek-distilled-qwen, Parameters=1.5B, LoRA=false2026.0215.7———
SLMBackbone=Mistral, Parameters=7B, LoRA=false2026.0215.1———
TCBackbone=Mistral, Parameters=7B, LoRA=true2026.0214.9———
TCBackbone=Mistral, Parameters=7B, LoRA=false2026.0214.8———
SLMBackbone=Mistral, Parameters=7B, LoRA=true2026.0214.2———
DABackbone=Mistral, Parameters=7B, LoRA=false2026.0214.1———
DFBackbone=Mistral, Parameters=7B, LoRA=true2026.0213.6———
NormalBackbone=Mistral, Parameters=7B, LoRA=false2026.0213.3———
×2 EpBackbone=Mistral, Parameters=7B, LoRA=false2026.0213.3———
×2 EpBackbone=Mistral, Parameters=7B, LoRA=true2026.0212.9———
DABackbone=Mistral, Parameters=7B, LoRA=true2026.0212.9———
NormalBackbone=Mistral, Parameters=7B, LoRA=true2026.0212.8———
DFBackbone=Mistral, Parameters=7B, LoRA=false2026.0212.4———
CABackbone=Deepseek-distilled-qwen, Parameters=1.5B, LoRA=false2026.0211.9———
SLMBackbone=Llama-3.1, Parameters=8B, LoRA=true2026.028.9———
CABackbone=Mistral, Parameters=7B, LoRA=false2026.028.7———
CABackbone=Mistral, Parameters=7B, LoRA=true2026.028.7———
TCBackbone=Llama-3.1, Parameters=8B, LoRA=true2026.028.6———
XTFBackbone=Llama-3.1, Parameters=8B, LoRA=false2026.028.4———
DABackbone=Llama-3.1, Parameters=8B, LoRA=false2026.028.2———
XTFBackbone=Llama-3.1, Parameters=8B, LoRA=true2026.028.1———
SLMBackbone=Llama-3.1, Parameters=8B, LoRA=false2026.027.9———
TCBackbone=Llama-3.1, Parameters=8B, LoRA=false2026.027.6———
×2 EpBackbone=Llama-3.1, Parameters=8B, LoRA=false2026.026.9———
NormalBackbone=Llama-3.1, Parameters=8B, LoRA=false2026.026.8———
×2 EpBackbone=Llama-3.1, Parameters=8B, LoRA=true2026.025.4———
NormalBackbone=Llama-3.1, Parameters=8B, LoRA=true2026.025.2———
DFBackbone=Llama-3.1, Parameters=8B, LoRA=false2026.025.1———
DABackbone=Llama-3.1, Parameters=8B, LoRA=true2026.024.8———
DFBackbone=Llama-3.1, Parameters=8B, LoRA=true2026.024.3———
XTFBackbone=Llama-3.2, Parameters=3B, LoRA=false2026.023.5———
TCBackbone=Llama-3.2, Parameters=3B, LoRA=false2026.023.4———
SLMBackbone=Llama-3.2, Parameters=3B, LoRA=false2026.023.2———
CABackbone=Llama-3.1, Parameters=8B, LoRA=false2026.022.5———
CABackbone=Llama-3.1, Parameters=8B, LoRA=true2026.022.5———
DFBackbone=Llama-3.2, Parameters=3B, LoRA=false2026.022.2———
DABackbone=Llama-3.2, Parameters=3B, LoRA=false2026.021.6———
NormalBackbone=Llama-3.2, Parameters=3B, LoRA=false2026.021.1———
×2 EpBackbone=Llama-3.2, Parameters=3B, LoRA=false2026.021———
DFBackbone=Llama-3.2, Parameters=1B, LoRA=false2026.020.6———
SLMBackbone=Llama-3.2, Parameters=1B, LoRA=false2026.020.5———
TCBackbone=Llama-3.2, Parameters=1B, LoRA=false2026.020.5———
XTFBackbone=Llama-3.2, Parameters=1B, LoRA=false2026.020.5———
CABackbone=Llama-3.2, Parameters=3B, LoRA=false2026.020.5———
NormalBackbone=Llama-3.2, Parameters=1B, LoRA=false2026.020.4———
×2 EpBackbone=Llama-3.2, Parameters=1B, LoRA=false2026.020.4———
DABackbone=Llama-3.2, Parameters=1B, LoRA=false2026.020.3———
CABackbone=Llama-3.2, Parameters=1B, LoRA=false2026.020———