# Multiple-choice Egocentric Question Answering on EgoCross CloseQA v1.0 (test) Source: [Multiple-choice Egocentric Question Answering on EgoCross CloseQA v1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-05-26T08:38:26.711Z Evaluation of model performance on the test split of the EgoCross CloseQA v1.0 dataset for Multiple-choice Egocentric Question Answering, including category-specific scores. 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 | Multiple-choice Egocentric Question Answering | | Dataset (including any stated split) | EgoCross CloseQA v1.0 (test) | | Metric | Animal Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [Animal Accuracy](): Higher is better; unit: PERCENT. - [XSports Accuracy](): Higher is better; unit: PERCENT. - [Industry Accuracy](): Higher is better; unit: PERCENT. - [Surgery Accuracy](): Higher is better; unit: PERCENT. - [Overall Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Animal Accuracy | Result date | Report paper | | --- | --- | --- | --- | | OmniEgo-R2 | 74.32 | Not recorded | 6a15292fdbd7150506f7dfbe | | OmniEgo-R2 | 74.32 | Not recorded | 6a15292fdbd7150506f7dfbe | | Gemini 2.5 Pro | 68.85 | Not recorded | 6a15292fdbd7150506f7dfbe | | GPT-4.1 | 64.48 | Not recorded | 6a15292fdbd7150506f7dfbe | | Qwen2.5-VL-7B | 53.55 | Not recorded | 6a15292fdbd7150506f7dfbe | | VideoLLaMA3-7B | 50.27 | Not recorded | 6a15292fdbd7150506f7dfbe | | InternVL3-8B | 49.18 | Not recorded | 6a15292fdbd7150506f7dfbe | | Qwen2.5-VL-3B | 41.53 | Not recorded | 6a15292fdbd7150506f7dfbe | | EgoGPT | 41.53 | Not recorded | 6a15292fdbd7150506f7dfbe | | EgoVLPv2 | 24.04 | Not recorded | 6a15292fdbd7150506f7dfbe | ### OmniEgo-R2 Animal Accuracy: 74.32. Result ID: 6a155b16fbc373592c0cc712. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Group":"Challenge Submissions","Track":"Source-Limited"}. This record does not assert independent reproduction. ### OmniEgo-R2 Animal Accuracy: 74.32. Result ID: 6a155b16fbc373592c0cc713. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Group":"Challenge Submissions","Track":"Open-Source"}. This record does not assert independent reproduction. ### Gemini 2.5 Pro Animal Accuracy: 68.85. Result ID: 6a155b16fbc373592c0cc70b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Group":"Proprietary MLLMs"}. This record does not assert independent reproduction. ### GPT-4.1 Animal Accuracy: 64.48. Result ID: 6a155b16fbc373592c0cc70a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Group":"Proprietary MLLMs"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Animal Accuracy: 53.55. Result ID: 6a155b16fbc373592c0cc70d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Group":"Open-source MLLMs"}. This record does not assert independent reproduction. ### VideoLLaMA3-7B Animal Accuracy: 50.27. Result ID: 6a155b16fbc373592c0cc70e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Group":"Open-source MLLMs"}. This record does not assert independent reproduction. ### InternVL3-8B Animal Accuracy: 49.18. Result ID: 6a155b16fbc373592c0cc70f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Group":"Open-source MLLMs"}. This record does not assert independent reproduction. ### Qwen2.5-VL-3B Animal Accuracy: 41.53. Result ID: 6a155b16fbc373592c0cc70c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Group":"Open-source MLLMs"}. This record does not assert independent reproduction. ### EgoGPT Animal Accuracy: 41.53. Result ID: 6a155b16fbc373592c0cc711. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Group":"Egocentric MLLMs"}. This record does not assert independent reproduction. ### EgoVLPv2 Animal Accuracy: 24.04. Result ID: 6a155b16fbc373592c0cc710. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Group":"Egocentric MLLMs"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.