# Multi-hop Question Answering on 2WikiMultiHopQA (official evaluation) Source: [Multi-hop Question Answering on 2WikiMultiHopQA (official evaluation)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:42:30.208Z Evaluation of model performance on the official evaluation set of the 2WikiMultiHopQA dataset for multi-hop question answering, typically measured using Exact Match (EM). 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 (2WikiMQA) (official evaluation) | | Metric | Exact Match (EM) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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. ## Recorded results | Method | Exact Match (EM) | Result date | Report paper | | --- | --- | --- | --- | | HRPO | 31.8 | Not recorded | 692e7315d6b0dd3342932746 | | M3PO | 31.4 | Not recorded | 692e7315d6b0dd3342932746 | | GRPO | 30.3 | Not recorded | 692e7315d6b0dd3342932746 | | PPO | 29.3 | Not recorded | 692e7315d6b0dd3342932746 | | M3PO | 27.9 | Not recorded | 692e7315d6b0dd3342932746 | | HRPO | 27.6 | Not recorded | 692e7315d6b0dd3342932746 | | QA | 25 | Not recorded | 692e7315d6b0dd3342932746 | | SFT | 24.8 | Not recorded | 692e7315d6b0dd3342932746 | | PPO | 24.2 | Not recorded | 692e7315d6b0dd3342932746 | | RAG | 23.5 | Not recorded | 692e7315d6b0dd3342932746 | | RAG | 22.6 | Not recorded | 692e7315d6b0dd3342932746 | | GRPO | 21.3 | Not recorded | 692e7315d6b0dd3342932746 | | SFT | 21 | Not recorded | 692e7315d6b0dd3342932746 | | RAG | 20.3 | Not recorded | 692e7315d6b0dd3342932746 | | Search-o1 | 17.6 | Not recorded | 692e7315d6b0dd3342932746 | | IRCoT | 14.9 | Not recorded | 692e7315d6b0dd3342932746 | | CoT | 11.1 | Not recorded | 692e7315d6b0dd3342932746 | ### HRPO Exact Match (EM): 31.8. Result ID: 699633a25698665ec475d5ef. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Strategy":"Hybrid Reasoning Policy Optimization","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### M3PO Exact Match (EM): 31.4. Result ID: 699633a25698665ec475d5f0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Strategy":"Multi-Path Perception Policy Optimization","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### GRPO Exact Match (EM): 30.3. Result ID: 699633a25698665ec475d5ee. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Strategy":"GRPO","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### PPO Exact Match (EM): 29.3. Result ID: 699633a25698665ec475d5ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Strategy":"PPO","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### M3PO Exact Match (EM): 27.9. Result ID: 699633a25698665ec475d5ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct","Strategy":"Multi-Path Perception Policy Optimization","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### HRPO Exact Match (EM): 27.6. Result ID: 699633a25698665ec475d5e9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct","Strategy":"Hybrid Reasoning Policy Optimization","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### QA Exact Match (EM): 25. Result ID: 699633a25698665ec475d5e0. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Strategy":"Direct inference","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### SFT Exact Match (EM): 24.8. Result ID: 699633a25698665ec475d5eb. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Strategy":"Supervised Fine-Tuning","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### PPO Exact Match (EM): 24.2. Result ID: 699633a25698665ec475d5e7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct","Strategy":"PPO","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### RAG Exact Match (EM): 23.5. Result ID: 699633a25698665ec475d5e4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Strategy":"Retrieval-Augmented Generation","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### RAG Exact Match (EM): 22.6. Result ID: 699633a25698665ec475d5ec. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Strategy":"Retrieval-Augmented Generation","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### GRPO Exact Match (EM): 21.3. Result ID: 699633a25698665ec475d5e8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct","Strategy":"GRPO","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### SFT Exact Match (EM): 21. Result ID: 699633a25698665ec475d5e5. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct","Strategy":"Supervised Fine-Tuning","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### RAG Exact Match (EM): 20.3. Result ID: 699633a25698665ec475d5e6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct","Strategy":"Retrieval-Augmented Generation","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### Search-o1 Exact Match (EM): 17.6. Result ID: 699633a25698665ec475d5e3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Strategy":"Search-o1","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### IRCoT Exact Match (EM): 14.9. Result ID: 699633a25698665ec475d5e2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Strategy":"Interleaved retrieval with CoT","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. ### CoT Exact Match (EM): 11.1. Result ID: 699633a25698665ec475d5e1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Strategy":"Chain-of-Thought","Context Retrieval":"Top-3 documents"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.