# Multihop Question Answering on HotpotQA (EM, ASR) Source: [Multihop Question Answering on HotpotQA (EM, ASR)]() Collection index: [llms.txt]() Data updated: 2026-07-02T08:40:29.004Z Evaluation of model performance on the HotpotQA dataset for multi-hop question answering, reporting Exact Match (EM) and Answer Selection Rate (ASR) scores. 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 | Multihop Question Answering | | Dataset (including any stated split) | HotpotQA | | Metric | EM | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 36 of 36 | | 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 - [EM](): Higher is better; unit: PERCENT. - [ASR](): Higher is better; unit: PERCENT. ## Recorded results | Method | EM | Result date | Report paper | | --- | --- | --- | --- | | Clean (No Attack) | 75 | Not recorded | 6a45f087dc71372d7795a25c | | Topic | 75 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 72 | Not recorded | 6a45f087dc71372d7795a25c | | Fake Completion | 71 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 71 | Not recorded | 6a45f087dc71372d7795a25c | | Fake Completion | 71 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 70 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 70 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 70 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 69 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 69 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 69 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 68 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 68 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 68 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 68 | Not recorded | 6a45f087dc71372d7795a25c | | Fake Completion | 67 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 67 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 66 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 66 | Not recorded | 6a45f087dc71372d7795a25c | | Topic | 64 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 58 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 57 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 56 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 55 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 55 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 52 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 50 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 40 | Not recorded | 6a45f087dc71372d7795a25c | | Topic | 40 | Not recorded | 6a45f087dc71372d7795a25c | | Fake Completion | 33 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 29 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 23 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 20 | Not recorded | 6a45f087dc71372d7795a25c | | Topic | 15 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 13 | Not recorded | 6a45f087dc71372d7795a25c | ### Clean (No Attack) EM: 75. Result ID: 6a4623fb60b88bb339854035. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### Topic EM: 75. Result ID: 6a4623fb60b88bb339854043. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Combined EM: 72. Result ID: 6a4623fb60b88bb33985404b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Fake Completion EM: 71. Result ID: 6a4623fb60b88bb339854038. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### Clean (No Attack) EM: 71. Result ID: 6a4623fb60b88bb33985403e. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Fake Completion EM: 71. Result ID: 6a4623fb60b88bb33985404a. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Naive EM: 70. Result ID: 6a4623fb60b88bb33985403f. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Ignore EM: 70. Result ID: 6a4623fb60b88bb339854040. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Combined EM: 70. Result ID: 6a4623fb60b88bb339854042. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Naive EM: 69. Result ID: 6a4623fb60b88bb339854036. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### Naive EM: 69. Result ID: 6a4623fb60b88bb339854048. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### PARADOX EM: 69. Result ID: 6a4623fb60b88bb33985404e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Clean (No Attack) EM: 68. Result ID: 6a4623fb60b88bb33985402c. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### PARADOX EM: 68. Result ID: 6a4623fb60b88bb33985403c. [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 EM: 68. Result ID: 6a4623fb60b88bb339854045. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Clean (No Attack) EM: 68. Result ID: 6a4623fb60b88bb339854047. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Fake Completion EM: 67. Result ID: 6a4623fb60b88bb339854041. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Ignore EM: 67. Result ID: 6a4623fb60b88bb339854049. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### PoisonedRAG EM: 66. Result ID: 6a4623fb60b88bb33985403b. [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 EM: 66. Result ID: 6a4623fb60b88bb339854044. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Topic EM: 64. Result ID: 6a4623fb60b88bb33985404c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### PoisonedRAG EM: 58. Result ID: 6a4623fb60b88bb33985404d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Naive EM: 57. Result ID: 6a4623fb60b88bb33985402d. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### Combined EM: 56. Result ID: 6a4623fb60b88bb339854039. [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 EM: 55. Result ID: 6a4623fb60b88bb33985402e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### PARADOX EM: 55. Result ID: 6a4623fb60b88bb339854033. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### KidnapRAG EM: 52. Result ID: 6a4623fb60b88bb33985404f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### PoisonedRAG EM: 50. Result ID: 6a4623fb60b88bb339854032. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### Ignore EM: 40. Result ID: 6a4623fb60b88bb339854037. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### Topic EM: 40. Result ID: 6a4623fb60b88bb33985403a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### Fake Completion EM: 33. Result ID: 6a4623fb60b88bb33985402f. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### Combined EM: 29. Result ID: 6a4623fb60b88bb339854030. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### KidnapRAG EM: 23. Result ID: 6a4623fb60b88bb339854046. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### KidnapRAG EM: 20. Result ID: 6a4623fb60b88bb339854034. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### Topic EM: 15. Result ID: 6a4623fb60b88bb339854031. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### KidnapRAG EM: 13. Result ID: 6a4623fb60b88bb33985403d. [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.