Multi-hop Question Answering on LB QA v1

77.3Accuracy

Qwen3-30B-A3B-Thinking-2507 + Data Recipe for Long-Context RL

Updated 1mo ago

Evaluation Results

Method	Links
Qwen3-30B-A3B-Thinking-2507 + Data Recipe for Long-Context RL 2026.06		77.3
Qwen3-30B-A3B-Thinking-2507 + DocQA-RL-1.6K 2026.06		68.9
Qwen3-8B-128K + Data Recipe for Long-Context RL 2026.06		68.8
Qwen3-30B-A3B-Thinking-2507 + KeyChain-15K 2026.06		68.4
Qwen3-4B-Thinking-2507 + Data Recipe for Long-Context RL 2026.06		68.3
Qwen3-4B-Thinking-2507 + KeyChain-15K 2026.06		66.8
Qwen3-8B-128K + KeyChain-15K 2026.06		66.5
Qwen3-30B-A3B-Thinking-2507 2026.06		66.5
Qwen3-4B-Thinking-2507 + DocQA-RL-1.6K 2026.06		63.8
Qwen3-8B-128K + DocQA-RL-1.6K 2026.06		63.7
Qwen3-8B-128K 2026.06		62.9
Qwen3-4B-Thinking-2507 2026.06		62.6