Our new X account is live! Follow @wizwand_team for updates

Knowledge-Intensive Reasoning on HotpotQA (F1 score)

0.654F1 Score

Llama3.1-8B + ARPO

Updated 2d ago

Evaluation Results

Method	Links
Llama3.1-8B + ARPO 2025.12		0.654
Qwen2.5-7B + GRPO 2025.12		0.59
Qwen2.5-7B + ARPO 2025.12		0.588
Qwen2.5-3B + ARPO 2025.12		0.585
Llama3.1-8B + GRPO 2025.12		0.578
Qwen2.5-7B + DAPO 2025.12		0.577
Llama3.1-8B + Reinforce ++ 2025.12		0.571
Llama3.1-8B + DAPO 2025.12		0.566
Qwen2.5-3B + GRPO 2025.12		0.565
Qwen2.5-3B + Reinforce ++ 2025.12		0.559
Qwen2.5-7B + Reinforce ++ 2025.12		0.551
Qwen2.5-3B + DAPO 2025.12		0.548
Llama3.1-8B + TIR Prompting 2025.12		0.485
Llama3.1-8B 2025.12		0.243
Qwen2.5-3B + TIR Prompting 2025.12		0.154
Qwen2.5-7B + TIR Prompting 2025.12		0.148
Qwen2.5-7B 2025.12		0.122
Qwen2.5-3B 2025.12		0.097