List your company

Coreference Resolution on SIMMC 2.1

59.98PrecisionQwen3
0.38815.85931.3346.801Apr 30, 2026

Evaluation results

MethodLinks
Qwen3Size=4B, Prompt setting=Reasoning2026.0459.9845.2151.55
Llama3.3Size=70B, Prompt setting=Reasoning2026.0458.7361.3760.02
DeepSeek-R1-Distill-Llama-8BSize=8B, Prompt setting=Reasoning2026.0453.543.9248.24
Llama3.1Size=8B, Prompt setting=Reasoning2026.0452.1659.2355.47
Qwen2.5Size=7B, Prompt setting=Reasoning2026.0449.957.3253.35
MistralSize=7B, Prompt setting=Reasoning2026.0439.8252.9345.45
Llama3.3Size=70B, Prompt setting=Few-shot2026.0439.3273.1251.14
Qwen3Size=4B, Prompt setting=Few-shot2026.0439.1359.1547.1
GemmaSize=7B, Prompt setting=Reasoning2026.0437.8643.2440.37
Llama3.1Size=8B, Prompt setting=Few-shot2026.0430.3468.3942.03
MistralSize=7B, Prompt setting=Few-shot2026.0429.1462.6539.77
Qwen3Size=4B, Prompt setting=Zero-shot2026.0429.0163.9139.9
Qwen2.5Size=7B, Prompt setting=Few-shot2026.0428.8470.0540.86
Llama3.1Size=8B, Prompt setting=Zero-shot2026.0427.7463.0338.53
Llama3.3Size=70B, Prompt setting=Zero-shot2026.0426.6180.139.95
MistralSize=7B, Prompt setting=Zero-shot2026.0425.766.6237.09
DeepSeek-R1-Distill-Llama-8BSize=8B, Prompt setting=Few-shot2026.0425.3167.1736.77
Qwen2.5Size=7B, Prompt setting=Zero-shot2026.0424.1471.6636.12
DeepSeek-R1-Distill-Llama-8BSize=8B, Prompt setting=Zero-shot2026.0422.9866.7734.19
GemmaSize=7B, Prompt setting=Few-shot2026.0419.5841.3826.58
GemmaSize=7B, Prompt setting=Zero-shot2026.0416.9352.7425.64
Random2026.042.6849.875.09