# Multi-hop Question Answering on LB QA v1 Source: [Multi-hop Question Answering on LB QA v1]() Collection index: [llms.txt]() Data updated: 2026-06-18T08:04:25.616Z Evaluation of model performance on the LB QA v1 dataset for multi-hop question answering, reporting Accuracy. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Multi-hop Question Answering | | Dataset (including any stated split) | LB QA v1 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Qwen3-30B-A3B-Thinking-2507 + Data Recipe for Long-Context RL | 77.3 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-30B-A3B-Thinking-2507 + DocQA-RL-1.6K | 68.9 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-8B-128K + Data Recipe for Long-Context RL | 68.8 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-30B-A3B-Thinking-2507 + KeyChain-15K | 68.4 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-4B-Thinking-2507 + Data Recipe for Long-Context RL | 68.3 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-4B-Thinking-2507 + KeyChain-15K | 66.8 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-8B-128K + KeyChain-15K | 66.5 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-30B-A3B-Thinking-2507 | 66.5 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-4B-Thinking-2507 + DocQA-RL-1.6K | 63.8 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-8B-128K + DocQA-RL-1.6K | 63.7 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-8B-128K | 62.9 | Not recorded | 6a337bc8206542251d7c523e | | Qwen3-4B-Thinking-2507 | 62.6 | Not recorded | 6a337bc8206542251d7c523e | ### Qwen3-30B-A3B-Thinking-2507 + Data Recipe for Long-Context RL Accuracy: 77.3. Result ID: 6a33a67ffbc373592c30c734. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-30B-A3B-Thinking-2507","Training Recipe":"Data Recipe for Long-Context RL"}. This record does not assert independent reproduction. ### Qwen3-30B-A3B-Thinking-2507 + DocQA-RL-1.6K Accuracy: 68.9. Result ID: 6a33a67ffbc373592c30c732. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-30B-A3B-Thinking-2507","Training Recipe":"DocQA-RL-1.6K"}. This record does not assert independent reproduction. ### Qwen3-8B-128K + Data Recipe for Long-Context RL Accuracy: 68.8. Result ID: 6a33a67ffbc373592c30c730. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-8B-128K","Training Recipe":"Data Recipe for Long-Context RL"}. This record does not assert independent reproduction. ### Qwen3-30B-A3B-Thinking-2507 + KeyChain-15K Accuracy: 68.4. Result ID: 6a33a67ffbc373592c30c733. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-30B-A3B-Thinking-2507","Training Recipe":"KeyChain-15K"}. This record does not assert independent reproduction. ### Qwen3-4B-Thinking-2507 + Data Recipe for Long-Context RL Accuracy: 68.3. Result ID: 6a33a67ffbc373592c30c72c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-4B-Thinking-2507","Training Recipe":"Data Recipe for Long-Context RL"}. This record does not assert independent reproduction. ### Qwen3-4B-Thinking-2507 + KeyChain-15K Accuracy: 66.8. Result ID: 6a33a67ffbc373592c30c72b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-4B-Thinking-2507","Training Recipe":"KeyChain-15K"}. This record does not assert independent reproduction. ### Qwen3-8B-128K + KeyChain-15K Accuracy: 66.5. Result ID: 6a33a67ffbc373592c30c72f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-8B-128K","Training Recipe":"KeyChain-15K"}. This record does not assert independent reproduction. ### Qwen3-30B-A3B-Thinking-2507 Accuracy: 66.5. Result ID: 6a33a67ffbc373592c30c731. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-30B-A3B-Thinking-2507","Training Recipe":"Base"}. This record does not assert independent reproduction. ### Qwen3-4B-Thinking-2507 + DocQA-RL-1.6K Accuracy: 63.8. Result ID: 6a33a67ffbc373592c30c72a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-4B-Thinking-2507","Training Recipe":"DocQA-RL-1.6K"}. This record does not assert independent reproduction. ### Qwen3-8B-128K + DocQA-RL-1.6K Accuracy: 63.7. Result ID: 6a33a67ffbc373592c30c72e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-8B-128K","Training Recipe":"DocQA-RL-1.6K"}. This record does not assert independent reproduction. ### Qwen3-8B-128K Accuracy: 62.9. Result ID: 6a33a67ffbc373592c30c72d. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-8B-128K","Training Recipe":"Base"}. This record does not assert independent reproduction. ### Qwen3-4B-Thinking-2507 Accuracy: 62.6. Result ID: 6a33a67ffbc373592c30c729. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-4B-Thinking-2507","Training Recipe":"Base"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.