# Multimodal Question Answering on MM-Lifelong week (test) Source: [Multimodal Question Answering on MM-Lifelong week (test)]() Collection index: [llms.txt]() Data updated: 2026-03-06T10:18:30.139Z Evaluation of model performance on the MM-Lifelong week dataset for Multimodal Question Answering tasks, typically measured by Acc and Ref@300. 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 | Multimodal Question Answering | | Dataset (including any stated split) | MM-Lifelong week (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 14 of 14 | | 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. - [Ref@300](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Human | 95.6 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | ReMA | 18.82 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Qwen3-VL-235B-A22B | 15.63 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | GPT-5 | 15 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Video-XL-2-8B | 12 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | VideoMind-7B | 11.75 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Qwen3-VL-30B-A3B | 11.07 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Nemotron-v2-12B | 11 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Video-XL-2-8B | 10.25 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | LongVT-7B | 9.75 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Eagle-2.5-8B | 9.5 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | DeepVideoDiscovery | 9.02 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Nemotron-v2-12B | 8.5 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Eagle-2.5-8B | 7 | Not recorded | 69aa5f9b2d603e6de3e41a87 | ### Human Accuracy: 95.6. Result ID: 69aaa9e96dbb7075b954f085. [Reporting paper]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### ReMA Accuracy: 18.82. Result ID: 69aaa9e96dbb7075b954f092. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### Qwen3-VL-235B-A22B Accuracy: 15.63. Result ID: 69aaa9e96dbb7075b954f087. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"1536"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 15. Result ID: 69aaa9e96dbb7075b954f086. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"50"}. This record does not assert independent reproduction. ### Video-XL-2-8B Accuracy: 12. Result ID: 69aaa9e96dbb7075b954f08a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"1024"}. This record does not assert independent reproduction. ### VideoMind-7B Accuracy: 11.75. Result ID: 69aaa9e96dbb7075b954f08f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### Qwen3-VL-30B-A3B Accuracy: 11.07. Result ID: 69aaa9e96dbb7075b954f088. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"1536"}. This record does not assert independent reproduction. ### Nemotron-v2-12B Accuracy: 11. Result ID: 69aaa9e96dbb7075b954f08d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"512"}. This record does not assert independent reproduction. ### Video-XL-2-8B Accuracy: 10.25. Result ID: 69aaa9e96dbb7075b954f089. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"2048"}. This record does not assert independent reproduction. ### LongVT-7B Accuracy: 9.75. Result ID: 69aaa9e96dbb7075b954f090. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### Eagle-2.5-8B Accuracy: 9.5. Result ID: 69aaa9e96dbb7075b954f08b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"512"}. This record does not assert independent reproduction. ### DeepVideoDiscovery Accuracy: 9.02. Result ID: 69aaa9e96dbb7075b954f091. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### Nemotron-v2-12B Accuracy: 8.5. Result ID: 69aaa9e96dbb7075b954f08e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"128"}. This record does not assert independent reproduction. ### Eagle-2.5-8B Accuracy: 7. Result ID: 69aaa9e96dbb7075b954f08c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"32"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.