# Video Reasoning Segmentation on ReVOS (test) Source: [Video Reasoning Segmentation on ReVOS (test)]() Collection index: [llms.txt]() Data updated: 2026-04-14T11:50:59.171Z Evaluation of Video Reasoning Segmentation performance on the test split of the ReVOS dataset, reporting detailed Jaccard (J), F-Measure (F), and J&F scores across referring, reasoning, and overall categories. 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 | Video Reasoning Segmentation | | Dataset (including any stated split) | ReVOS (test) | | Metric | Referring Jaccard | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 19 of 19 | | 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 - [Referring Jaccard](): Higher is better; unit: PERCENT. - [Referring F-Measure](): Higher is better; unit: PERCENT. - [Referring J&F](): Higher is better; unit: PERCENT. - [Reasoning Jaccard](): Higher is better; unit: PERCENT. - [Reasoning F-Measure](): Higher is better; unit: PERCENT. - [Reasoning J&F](): Higher is better; unit: PERCENT. - [Overall Jaccard](): Higher is better; unit: PERCENT. - [Overall F-Measure](): Higher is better; unit: PERCENT. - [Overall J&F](): Higher is better; unit: PERCENT. - [R Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Referring Jaccard | Result date | Report paper | | --- | --- | --- | --- | | VRS-HQ | 61.1 | Not recorded | 68fa77271e209131cea7168d | | VRS-HQ | 59.8 | Not recorded | 68fa77271e209131cea7168d | | VRS-HQ | 59.8 | Not recorded | 69ddca4f7d0172140ec79d31 | | GLUS | 56.3 | Not recorded | 69ddca4f7d0172140ec79d31 | | VISA | 55.7 | Not recorded | 68fa77271e209131cea7168d | | ORVOS Baseline | 55.7 | Not recorded | 69ddca4f7d0172140ec79d31 | | VISA | 55.6 | Not recorded | 68fa77271e209131cea7168d | | VISA | 49.4 | Not recorded | 68fa77271e209131cea7168d | | VISA | 49.2 | Not recorded | 68fa77271e209131cea7168d | | VISA | 49.2 | Not recorded | 69ddca4f7d0172140ec79d31 | | TrackGPT | 48.3 | Not recorded | 68fa77271e209131cea7168d | | TrackGPT | 46.7 | Not recorded | 68fa77271e209131cea7168d | | TrackGPT | 46.7 | Not recorded | 69ddca4f7d0172140ec79d31 | | LISA | 45.2 | Not recorded | 68fa77271e209131cea7168d | | LISA | 44.3 | Not recorded | 68fa77271e209131cea7168d | | LISA | 44.3 | Not recorded | 69ddca4f7d0172140ec79d31 | | ReferFormer | 31.2 | Not recorded | 68fa77271e209131cea7168d | | MTTR | 29.8 | Not recorded | 68fa77271e209131cea7168d | | LMPM | 29 | Not recorded | 68fa77271e209131cea7168d | ### VRS-HQ Referring Jaccard: 61.1. Result ID: 699476233a6c3148491d6626. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Chat-UniVi-13B"}. This record does not assert independent reproduction. ### VRS-HQ Referring Jaccard: 59.8. Result ID: 699476233a6c3148491d6625. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Chat-UniVi-7B"}. This record does not assert independent reproduction. ### VRS-HQ Referring Jaccard: 59.8. Result ID: 69de2a1247f34ce924c6c962. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GLUS Referring Jaccard: 56.3. Result ID: 69de2a1247f34ce924c6c961. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### VISA Referring Jaccard: 55.7. Result ID: 699476233a6c3148491d6622. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-13B"}. This record does not assert independent reproduction. ### ORVOS Baseline Referring Jaccard: 55.7. Result ID: 69de2a1247f34ce924c6c963. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### VISA Referring Jaccard: 55.6. Result ID: 699476233a6c3148491d6624. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Chat-UniVi-13B"}. This record does not assert independent reproduction. ### VISA Referring Jaccard: 49.4. Result ID: 699476233a6c3148491d6621. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-7B"}. This record does not assert independent reproduction. ### VISA Referring Jaccard: 49.2. Result ID: 699476233a6c3148491d6623. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Chat-UniVi-7B"}. This record does not assert independent reproduction. ### VISA Referring Jaccard: 49.2. Result ID: 69de2a1247f34ce924c6c960. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### TrackGPT Referring Jaccard: 48.3. Result ID: 699476233a6c3148491d6620. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-13B"}. This record does not assert independent reproduction. ### TrackGPT Referring Jaccard: 46.7. Result ID: 699476233a6c3148491d661f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-7B"}. This record does not assert independent reproduction. ### TrackGPT Referring Jaccard: 46.7. Result ID: 69de2a1247f34ce924c6c95f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LISA Referring Jaccard: 45.2. Result ID: 699476233a6c3148491d661e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-13B"}. This record does not assert independent reproduction. ### LISA Referring Jaccard: 44.3. Result ID: 699476233a6c3148491d661d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA 7B"}. This record does not assert independent reproduction. ### LISA Referring Jaccard: 44.3. Result ID: 69de2a1247f34ce924c6c95e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ReferFormer Referring Jaccard: 31.2. Result ID: 699476233a6c3148491d661c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Video-Swin-B"}. This record does not assert independent reproduction. ### MTTR Referring Jaccard: 29.8. Result ID: 699476233a6c3148491d661a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Video-Swin-T"}. This record does not assert independent reproduction. ### LMPM Referring Jaccard: 29. Result ID: 699476233a6c3148491d661b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Swin-T"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.