# Multi-hop Question Answering on MMhops Comparison (test) Source: [Multi-hop Question Answering on MMhops Comparison (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:52.640Z Evaluation of model performance on the test split of the MMhops Comparison dataset for multi-hop question answering, reporting Accuracy. 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) | MMhops Comparison (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 13 of 13 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Gemini-2.5-pro | 29.39 | Not recorded | 6940e7e1456146936284fc83 | | Gemini-2.5-flash | 23.18 | Not recorded | 6940e7e1456146936284fc83 | | MMhops-R1 | 22.01 | Not recorded | 6940e7e1456146936284fc83 | | Self-Ask | 18.27 | Not recorded | 6940e7e1456146936284fc83 | | OmniSearch | 17.02 | Not recorded | 6940e7e1456146936284fc83 | | Vanilla mRAG | 9.72 | Not recorded | 6940e7e1456146936284fc83 | | GPT-4o | 8.76 | Not recorded | 6940e7e1456146936284fc83 | | Zero-shot | 7.59 | Not recorded | 6940e7e1456146936284fc83 | | GPT-4o-mini | 7.05 | Not recorded | 6940e7e1456146936284fc83 | | Search-r1 | 6.62 | Not recorded | 6940e7e1456146936284fc83 | | Zero-shot | 6.2 | Not recorded | 6940e7e1456146936284fc83 | | EchoSight | 4.81 | Not recorded | 6940e7e1456146936284fc83 | | Vanilla mRAG | 3.95 | Not recorded | 6940e7e1456146936284fc83 | ### Gemini-2.5-pro Accuracy: 29.39. Result ID: 699845af5698665ec4a265bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":null,"Retriever":null}. This record does not assert independent reproduction. ### Gemini-2.5-flash Accuracy: 23.18. Result ID: 699845af5698665ec4a265be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":null,"Retriever":null}. This record does not assert independent reproduction. ### MMhops-R1 Accuracy: 22.01. Result ID: 699845af5698665ec4a265c8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-vl-7B-Instruct","Retriever":"Image, Text"}. This record does not assert independent reproduction. ### Self-Ask Accuracy: 18.27. Result ID: 699845af5698665ec4a265c3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"GPT-4o","Retriever":"Caption, Text"}. This record does not assert independent reproduction. ### OmniSearch Accuracy: 17.02. Result ID: 699845af5698665ec4a265c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"GPT-4o","Retriever":"Image, Text"}. This record does not assert independent reproduction. ### Vanilla mRAG Accuracy: 9.72. Result ID: 699845af5698665ec4a265c5. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-vl-7B-Instruct","Retriever":"Image, Text"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 8.76. Result ID: 699845af5698665ec4a265bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":null,"Retriever":null}. This record does not assert independent reproduction. ### Zero-shot Accuracy: 7.59. Result ID: 699845af5698665ec4a265c1. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-vl-72B-Instruct","Retriever":null}. This record does not assert independent reproduction. ### GPT-4o-mini Accuracy: 7.05. Result ID: 699845af5698665ec4a265bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":null,"Retriever":null}. This record does not assert independent reproduction. ### Search-r1 Accuracy: 6.62. Result ID: 699845af5698665ec4a265c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7b-Instruct","Retriever":"Caption, Text"}. This record does not assert independent reproduction. ### Zero-shot Accuracy: 6.2. Result ID: 699845af5698665ec4a265c0. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-vl-7B-Instruct","Retriever":null}. This record does not assert independent reproduction. ### EchoSight Accuracy: 4.81. Result ID: 699845af5698665ec4a265c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"LLaMA3","Retriever":"Image, Text"}. This record does not assert independent reproduction. ### Vanilla mRAG Accuracy: 3.95. Result ID: 699845af5698665ec4a265c4. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-vl-7B-Instruct","Retriever":"Text"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.