# Multi-hop Question Answering on MuSiQue (speedup) Source: [Multi-hop Question Answering on MuSiQue (speedup)]() Collection index: [llms.txt]() Data updated: 2026-06-02T10:16:38.271Z Evaluation of model efficiency on the MuSiQue dataset for Multi-hop Question Answering, specifically reporting the speedup factor. 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) | MuSiQue | | Metric | Speedup | | Unit | SCALAR | | 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 - [Speedup](): Higher is better; unit: SCALAR. ## Recorded results | Method | Speedup | Result date | Report paper | | --- | --- | --- | --- | | Hybrid Verified Decoding | 3.55 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 3.22 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 3 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 2.88 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 2.12 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 1.97 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 1.35 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 1.18 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 1.17 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 1.15 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 1.05 | Not recorded | 6a1e63fadbd7150506f85546 | | Hybrid Verified Decoding | 0.95 | Not recorded | 6a1e63fadbd7150506f85546 | ### Hybrid Verified Decoding Speedup: 3.55. Result ID: 6a1ead43fbc373592c196af9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-8B","Reference Baseline":"Greedy Decoding"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 3.22. Result ID: 6a1ead43fbc373592c196b02. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Llama3.1-8B","Reference Baseline":"EAGLE3"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 3. Result ID: 6a1ead43fbc373592c196b01. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Llama3.1-8B","Reference Baseline":"Greedy Decoding"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 2.88. Result ID: 6a1ead43fbc373592c196afb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-8B","Reference Baseline":"SuffixDecoding"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 2.12. Result ID: 6a1ead43fbc373592c196afd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-4B","Reference Baseline":"Greedy Decoding"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 1.97. Result ID: 6a1ead43fbc373592c196b04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Llama3.1-8B","Reference Baseline":"Frequency-Based Hybrid Decoding"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 1.35. Result ID: 6a1ead43fbc373592c196aff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-4B","Reference Baseline":"SuffixDecoding"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 1.18. Result ID: 6a1ead43fbc373592c196afe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-4B","Reference Baseline":"EAGLE3"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 1.17. Result ID: 6a1ead43fbc373592c196afc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-8B","Reference Baseline":"Frequency-Based Hybrid Decoding"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 1.15. Result ID: 6a1ead43fbc373592c196b03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Llama3.1-8B","Reference Baseline":"SuffixDecoding"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 1.05. Result ID: 6a1ead43fbc373592c196b00. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-4B","Reference Baseline":"Frequency-Based Hybrid Decoding"}. This record does not assert independent reproduction. ### Hybrid Verified Decoding Speedup: 0.95. Result ID: 6a1ead43fbc373592c196afa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-8B","Reference Baseline":"EAGLE3"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.