# Multi-hop Question Answering on 2WikiMultihopQA (EM, ASR) Source: [Multi-hop Question Answering on 2WikiMultihopQA (EM, ASR)]() Collection index: [llms.txt]() Data updated: 2026-07-02T08:40:29.192Z Evaluation of model performance on the 2WikiMultihopQA dataset for multi-hop question answering, reporting Exact Match (EM) and Answer Success 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) | 2WikiMultihopQA | | Metric | Exact Match (EM) | | Unit | SCALAR | | 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: SCALAR. - [Answer Success Rate (ASR)](): Higher is better; unit: SCALAR. ## Recorded results | Method | Exact Match (EM) | Result date | Report paper | | --- | --- | --- | --- | | Clean (No Attack) | 0.6 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 0.59 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 0.59 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 0.58 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 0.58 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 0.57 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 0.56 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 0.55 | Not recorded | 6a45f087dc71372d7795a25c | | Clean (No Attack) | 0.54 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 0.53 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 0.51 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 0.47 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 0.41 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 0.41 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 0.38 | Not recorded | 6a45f087dc71372d7795a25c | | Naive | 0.37 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 0.37 | Not recorded | 6a45f087dc71372d7795a25c | | PARADOX | 0.36 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 0.35 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 0.34 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 0.34 | Not recorded | 6a45f087dc71372d7795a25c | | PoisonedRAG | 0.34 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 0.29 | Not recorded | 6a45f087dc71372d7795a25c | | Combined | 0.26 | Not recorded | 6a45f087dc71372d7795a25c | | Ignore | 0.24 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 0.2 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 0.09 | Not recorded | 6a45f087dc71372d7795a25c | | KidnapRAG | 0.09 | Not recorded | 6a45f087dc71372d7795a25c | ### Clean (No Attack) Exact Match (EM): 0.6. Result ID: 6a4623fb60b88bb339854130. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### Ignore Exact Match (EM): 0.59. Result ID: 6a4623fb60b88bb339854140. [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): 0.59. Result ID: 6a4623fb60b88bb339854147. [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): 0.58. Result ID: 6a4623fb60b88bb339854145. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Combined Exact Match (EM): 0.58. Result ID: 6a4623fb60b88bb339854148. [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): 0.57. Result ID: 6a4623fb60b88bb33985413e. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Naive Exact Match (EM): 0.56. Result ID: 6a4623fb60b88bb33985413f. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Naive Exact Match (EM): 0.55. Result ID: 6a4623fb60b88bb339854146. [Reporting paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### Clean (No Attack) Exact Match (EM): 0.54. Result ID: 6a4623fb60b88bb339854137. [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): 0.53. Result ID: 6a4623fb60b88bb339854141. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### PARADOX Exact Match (EM): 0.51. Result ID: 6a4623fb60b88bb339854143. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### Naive Exact Match (EM): 0.47. Result ID: 6a4623fb60b88bb339854138. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### PoisonedRAG Exact Match (EM): 0.41. Result ID: 6a4623fb60b88bb33985413b. [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): 0.41. Result ID: 6a4623fb60b88bb33985413c. [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): 0.38. Result ID: 6a4623fb60b88bb33985414a. [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): 0.37. Result ID: 6a4623fb60b88bb339854131. [Reporting paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### PoisonedRAG Exact Match (EM): 0.37. Result ID: 6a4623fb60b88bb339854134. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### PARADOX Exact Match (EM): 0.36. Result ID: 6a4623fb60b88bb339854135. [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): 0.35. Result ID: 6a4623fb60b88bb339854149. [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): 0.34. Result ID: 6a4623fb60b88bb339854132. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### Combined Exact Match (EM): 0.34. Result ID: 6a4623fb60b88bb33985413a. [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): 0.34. Result ID: 6a4623fb60b88bb339854142. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### KidnapRAG Exact Match (EM): 0.29. Result ID: 6a4623fb60b88bb339854144. [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): 0.26. Result ID: 6a4623fb60b88bb339854133. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Qwen2.5-32B-Inst"}. This record does not assert independent reproduction. ### Ignore Exact Match (EM): 0.24. Result ID: 6a4623fb60b88bb339854139. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"ReAct","LLM Backbone":"Llama-3.3-70B-Inst"}. This record does not assert independent reproduction. ### KidnapRAG Exact Match (EM): 0.2. Result ID: 6a4623fb60b88bb33985414b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"WebThinker","LLM Backbone":"DeepSeek-32B"}. This record does not assert independent reproduction. ### KidnapRAG Exact Match (EM): 0.09. Result ID: 6a4623fb60b88bb339854136. [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): 0.09. Result ID: 6a4623fb60b88bb33985413d. [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.