# Referring Expression Comprehension on GRIT refexp Source: [Referring Expression Comprehension on GRIT refexp]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:20.287Z Evaluation of Referring Expression Comprehension performance on the GRIT refexp dataset, typically measured using 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 | Referring Expression Comprehension | | Dataset (including any stated split) | GRIT refexp | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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 | | --- | --- | --- | --- | | Unified-IO | 78.61 | Not recorded | 68fa70d41e209131cea4d774 | | Unified-IO | 78.6 | Not recorded | 68fa6fe61e209131cea47fbc | | Qwen-VL-7B | 78.22 | Not recorded | 68fa70d41e209131cea4d774 | | Shikra-7B | 69.34 | Not recorded | 68fa6fe61e209131cea47fbc | | Shikra-7B | 69.34 | Not recorded | 68fa70d41e209131cea4d774 | | Upper Bound | 69.34 | Not recorded | 68fa773e1e209131cea71ed4 | | Shikra-13B | 69.03 | Not recorded | 68fa6fe61e209131cea47fbc | | Shikra-13B | 69.03 | Not recorded | 68fa70d41e209131cea4d774 | | VoCo-LLaMA | 68.59 | Not recorded | 68fa773e1e209131cea71ed4 | | VoCo-LLaMA | 67.69 | Not recorded | 68fa773e1e209131cea71ed4 | | OFA-L\* | 61.7 | Not recorded | 68fa6fe61e209131cea47fbc | | OFA-L | 61.7 | Not recorded | 68fa70d41e209131cea4d774 | | Lower Bound | 60.5 | Not recorded | 68fa773e1e209131cea71ed4 | | GPV-2 | 51.5 | Not recorded | 68fa6fe61e209131cea47fbc | | GPV-2 | 51.5 | Not recorded | 68fa70d41e209131cea4d774 | ### Unified-IO Accuracy: 78.61. Result ID: 69915b843a6c314849079693. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model type":"Generalist Models"}. This record does not assert independent reproduction. ### Unified-IO Accuracy: 78.6. Result ID: 6990f0d23a6c314849047469. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model type":"Generalist VL SOTAs (w/o finetuning)"}. This record does not assert independent reproduction. ### Qwen-VL-7B Accuracy: 78.22. Result ID: 69915b843a6c314849079696. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model type":"Generalist Models"}. This record does not assert independent reproduction. ### Shikra-7B Accuracy: 69.34. Result ID: 6990f0d23a6c31484904746a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model type":"Generalist VL SOTAs (w/o finetuning)"}. This record does not assert independent reproduction. ### Shikra-7B Accuracy: 69.34. Result ID: 69915b843a6c314849079694. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model type":"Generalist Models"}. This record does not assert independent reproduction. ### Upper Bound Accuracy: 69.34. Result ID: 69946f443a6c3148491d3877. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Token":"256"}. This record does not assert independent reproduction. ### Shikra-13B Accuracy: 69.03. Result ID: 6990f0d23a6c31484904746b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model type":"Generalist VL SOTAs (w/o finetuning)"}. This record does not assert independent reproduction. ### Shikra-13B Accuracy: 69.03. Result ID: 69915b843a6c314849079695. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model type":"Generalist Models"}. This record does not assert independent reproduction. ### VoCo-LLaMA Accuracy: 68.59. Result ID: 69946f443a6c3148491d3878. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Token":"8"}. This record does not assert independent reproduction. ### VoCo-LLaMA Accuracy: 67.69. Result ID: 69946f443a6c3148491d3879. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Token":"1"}. This record does not assert independent reproduction. ### OFA-L\* Accuracy: 61.7. Result ID: 6990f0d23a6c314849047468. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model type":"Generalist VL SOTAs (w/o finetuning)"}. This record does not assert independent reproduction. ### OFA-L Accuracy: 61.7. Result ID: 69915b843a6c314849079692. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model type":"Generalist Models"}. This record does not assert independent reproduction. ### Lower Bound Accuracy: 60.5. Result ID: 69946f443a6c3148491d387a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Token":"1"}. This record does not assert independent reproduction. ### GPV-2 Accuracy: 51.5. Result ID: 6990f0d23a6c314849047467. [Reporting paper]() Recorded evaluation configuration: {"Model type":"Generalist VL SOTAs (w/o finetuning)"}. This record does not assert independent reproduction. ### GPV-2 Accuracy: 51.5. Result ID: 69915b843a6c314849079691. [Reporting paper]() Recorded evaluation configuration: {"Model type":"Generalist Models"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.