# Multi-hop Question Answering on HotpotQA (EM, ASR) Source: [Multi-hop Question Answering on HotpotQA (EM, ASR)]() Collection index: [llms.txt]() Data updated: 2026-07-02T08:40:29.147Z Evaluation of model performance on the HotpotQA dataset for multi-hop question answering, reporting Exact Match (EM) and Answer Selection Rate (ASR). 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) | HotpotQA | | Metric | Exact Match (EM) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 28 of 28 | | 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 - [Exact Match (EM)](): Higher is better; unit: PERCENT. - [Answer Selection Rate (ASR)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Exact Match (EM) | Result date | Report paper | | --- | --- | --- | --- | | Ignore | 76 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 75 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 75 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 71 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 70 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 70 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 69 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 69 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 68 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 68 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 67 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 66 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 64 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 64 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 62 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 57 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 57 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 56 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 56 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 50 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 48 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 47 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 46 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 44 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 35 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 34 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 20 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 20 | Not recorded | 6a45f087dc71372d7795a25c | ### Ignore Exact Match (EM): 76. Result ID: 6a4623fb60b88bb339854108. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Clean (No Attack) Exact Match (EM): 75. Result ID: 6a4623fb60b88bb3398540ff. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### Combined Exact Match (EM): 75. Result ID: 6a4623fb60b88bb339854109. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Clean (No Attack) Exact Match (EM): 71. Result ID: 6a4623fb60b88bb339854106. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Naive Exact Match (EM): 70. Result ID: 6a4623fb60b88bb339854107. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### PARADOX Exact Match (EM): 70. Result ID: 6a4623fb60b88bb33985410b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Ignore Exact Match (EM): 69. Result ID: 6a4623fb60b88bb33985410f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Combined Exact Match (EM): 69. Result ID: 6a4623fb60b88bb339854110. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Clean (No Attack) Exact Match (EM): 68. Result ID: 6a4623fb60b88bb3398540f8. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### Clean (No Attack) Exact Match (EM): 68. Result ID: 6a4623fb60b88bb33985410d. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Naive Exact Match (EM): 67. Result ID: 6a4623fb60b88bb339854100. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### Naive Exact Match (EM): 66. Result ID: 6a4623fb60b88bb33985410e. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### PARADOX Exact Match (EM): 64. Result ID: 6a4623fb60b88bb339854104. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### PoisonedRAG Exact Match (EM): 64. Result ID: 6a4623fb60b88bb33985410a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Combined Exact Match (EM): 62. Result ID: 6a4623fb60b88bb339854102. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### PARADOX Exact Match (EM): 57. Result ID: 6a4623fb60b88bb3398540fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### PoisonedRAG Exact Match (EM): 57. Result ID: 6a4623fb60b88bb339854111. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Naive Exact Match (EM): 56. Result ID: 6a4623fb60b88bb3398540f9. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### PoisonedRAG Exact Match (EM): 56. Result ID: 6a4623fb60b88bb339854103. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### PARADOX Exact Match (EM): 50. Result ID: 6a4623fb60b88bb339854112. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Ignore Exact Match (EM): 48. Result ID: 6a4623fb60b88bb339854101. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### Ignore Exact Match (EM): 47. Result ID: 6a4623fb60b88bb3398540fa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### KidnapRAG Exact Match (EM): 46. Result ID: 6a4623fb60b88bb339854113. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### PoisonedRAG Exact Match (EM): 44. Result ID: 6a4623fb60b88bb3398540fc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### KidnapRAG Exact Match (EM): 35. Result ID: 6a4623fb60b88bb33985410c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Combined Exact Match (EM): 34. Result ID: 6a4623fb60b88bb3398540fb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### KidnapRAG Exact Match (EM): 20. Result ID: 6a4623fb60b88bb3398540fe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### KidnapRAG Exact Match (EM): 20. Result ID: 6a4623fb60b88bb339854105. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.