# Multimodal Question Answering on MM-Lifelong (val@month) Source: [Multimodal Question Answering on MM-Lifelong (val@month)]() Collection index: [llms.txt]() Data updated: 2026-03-06T10:18:30.110Z Evaluation of model performance on the val@month split of the MM-Lifelong dataset for Multimodal Question Answering tasks, using metrics like Accuracy 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 (val@month) | | 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: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Human | 80.4 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | ReMA | 18.62 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | GPT-5 | 14.87 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Qwen3-VL-235B-A22B | 14.33 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Qwen3-VL-30B-A3B | 11.92 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | DeepVideoDiscovery | 10.57 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Nemotron-v2-12B | 10.03 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Nemotron-v2-12B | 9.63 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Video-XL-2-8B | 9.07 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Video-XL-2-8B | 8.91 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | VideoMind-7B | 8.35 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | LongVT-7B | 7.54 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Eagle-2.5-8B | 6.1 | Not recorded | 69aa5f9b2d603e6de3e41a87 | | Eagle-2.5-8B | 4.41 | Not recorded | 69aa5f9b2d603e6de3e41a87 | ### Human Accuracy: 80.4. Result ID: 69aaa9e96dbb7075b954f077. [Reporting paper]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### ReMA Accuracy: 18.62. Result ID: 69aaa9e96dbb7075b954f084. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 14.87. Result ID: 69aaa9e96dbb7075b954f078. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"50"}. This record does not assert independent reproduction. ### Qwen3-VL-235B-A22B Accuracy: 14.33. Result ID: 69aaa9e96dbb7075b954f079. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"1536"}. This record does not assert independent reproduction. ### Qwen3-VL-30B-A3B Accuracy: 11.92. Result ID: 69aaa9e96dbb7075b954f07a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"1536"}. This record does not assert independent reproduction. ### DeepVideoDiscovery Accuracy: 10.57. Result ID: 69aaa9e96dbb7075b954f083. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### Nemotron-v2-12B Accuracy: 10.03. Result ID: 69aaa9e96dbb7075b954f080. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"128"}. This record does not assert independent reproduction. ### Nemotron-v2-12B Accuracy: 9.63. Result ID: 69aaa9e96dbb7075b954f07f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"512"}. This record does not assert independent reproduction. ### Video-XL-2-8B Accuracy: 9.07. Result ID: 69aaa9e96dbb7075b954f07c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"1024"}. This record does not assert independent reproduction. ### Video-XL-2-8B Accuracy: 8.91. Result ID: 69aaa9e96dbb7075b954f07b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"2048"}. This record does not assert independent reproduction. ### VideoMind-7B Accuracy: 8.35. Result ID: 69aaa9e96dbb7075b954f081. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### LongVT-7B Accuracy: 7.54. Result ID: 69aaa9e96dbb7075b954f082. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Frames":"Full"}. This record does not assert independent reproduction. ### Eagle-2.5-8B Accuracy: 6.1. Result ID: 69aaa9e96dbb7075b954f07e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"32"}. This record does not assert independent reproduction. ### Eagle-2.5-8B Accuracy: 4.41. Result ID: 69aaa9e96dbb7075b954f07d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Frames":"512"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.