Temporal Question Answering on TransientTables (test) benchmark leaderboard · SOTA2 Research

Temporal Question Answering on TransientTables (test)

80.39EMGemini-2.0-Flash

Evaluation Results

Method	Links
Gemini-2.0-Flash2025.11		80.39	82.11
Llama-3.3-70B-Instruct2025.11		79.52	79.91
Llama-3.1-8B-Instruct2025.11		78.7	77.85
GPT-4o-mini2025.11		78.54	78.7
Gemini-2.5-Pro2025.11		78.51	78.2
Llama-3.1-8B-Instruct2025.11		78.08	79.49
Qwen2.5-7B-Instruct2025.11		77.84	78.86
Qwen2.5-7B-Instruct2025.11		75.3	75.07
Gemini-2.0-Flash2025.11		72.7	73.33
Llama-3.3-70B-Instruct2025.11		69.86	75.54
GPT-4o-mini2025.11		69.56	66.8
Gemini-2.5-Pro2025.11		68.89	75.3
Gemini-2.5-Pro2025.11		68.25	67.18
Llama-3.3-70B-Instruct2025.11		64.79	63.66
Gemini-2.5-Pro2025.11		63.13	63.07
GPT-4o-mini2025.11		62.88	63.15
Llama-3.1-8B-Instruct2025.11		61.78	60.66
Qwen2.5-7B-Instruct2025.11		60.91	60.32
Gemini-2.0-Flash2025.11		60.37	60.32
Gemini-2.0-Flash2025.11		48.98	55.93
GPT-4o-mini2025.11		44.53	52.27
Llama-3.3-70B-Instruct2025.11		41.08	54.91
Qwen2.5-7B-Instruct2025.11		30.4	29.22
Llama-3.1-8B-Instruct2025.11		29.83	37.95