# Referring Text Classification on COCO-Text Out-of-Domain Source: [Referring Text Classification on COCO-Text Out-of-Domain]() Collection index: [llms.txt]() Data updated: 2026-02-27T14:24:07.535Z Evaluation of Referring Text Classification performance on the Out-of-Domain split of the COCO-Text dataset, assessed 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 Text Classification | | Dataset (including any stated split) | COCO-Text Out-of-Domain | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 | | --- | --- | --- | --- | | LLaVA-1.5 + Blur | 84.62 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 + Blur | 75.85 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 + ControlMLLM++ | 74.66 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 + ControlMLLM++ | 74.65 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 + ControlMLLM | 63.06 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 + ControlMLLM | 61.43 | Not recorded | 699d30c56017c4460ffab7ec | | Shikra-7B | 60.41 | Not recorded | 699d30c56017c4460ffab7ec | | Ferret-7B | 59.81 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 + Color | 59.04 | Not recorded | 699d30c56017c4460ffab7ec | | Ferret-7B | 58.7 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 | 57.42 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 | 57.42 | Not recorded | 699d30c56017c4460ffab7ec | | GPT4-ROI | 57.17 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 + Color | 56.57 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 + Edit Att | 51.88 | Not recorded | 699d30c56017c4460ffab7ec | | LLaVA-1.5 + Edit Att | 50 | Not recorded | 699d30c56017c4460ffab7ec | | Kosmos-2 | 23.12 | Not recorded | 699d30c56017c4460ffab7ec | ### LLaVA-1.5 + Blur Accuracy: 84.62. Result ID: 69a1a90535dfcabb00e81088. [Reporting paper]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 + Blur Accuracy: 75.85. Result ID: 69a1a90535dfcabb00e81089. [Reporting paper]() Recorded evaluation configuration: {"Input format":"Mask","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 + ControlMLLM++ Accuracy: 74.66. Result ID: 69a1a90535dfcabb00e81090. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 + ControlMLLM++ Accuracy: 74.65. Result ID: 69a1a90535dfcabb00e81091. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input format":"Mask","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 + ControlMLLM Accuracy: 63.06. Result ID: 69a1a90535dfcabb00e8108e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 + ControlMLLM Accuracy: 61.43. Result ID: 69a1a90535dfcabb00e8108f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input format":"Mask","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### Shikra-7B Accuracy: 60.41. Result ID: 69a1a90535dfcabb00e81083. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training Methods"}. This record does not assert independent reproduction. ### Ferret-7B Accuracy: 59.81. Result ID: 69a1a90535dfcabb00e81085. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input format":"Mask","Training strategy":"Training Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 + Color Accuracy: 59.04. Result ID: 69a1a90535dfcabb00e8108a. [Reporting paper]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### Ferret-7B Accuracy: 58.7. Result ID: 69a1a90535dfcabb00e81084. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 Accuracy: 57.42. Result ID: 69a1a90535dfcabb00e81086. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 Accuracy: 57.42. Result ID: 69a1a90535dfcabb00e81087. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input format":"Mask","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### GPT4-ROI Accuracy: 57.17. Result ID: 69a1a90535dfcabb00e81082. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 + Color Accuracy: 56.57. Result ID: 69a1a90535dfcabb00e8108b. [Reporting paper]() Recorded evaluation configuration: {"Input format":"Mask","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 + Edit Att Accuracy: 51.88. Result ID: 69a1a90535dfcabb00e8108d. [Reporting paper]() Recorded evaluation configuration: {"Input format":"Mask","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### LLaVA-1.5 + Edit Att Accuracy: 50. Result ID: 69a1a90535dfcabb00e8108c. [Reporting paper]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training-Free Methods"}. This record does not assert independent reproduction. ### Kosmos-2 Accuracy: 23.12. Result ID: 69a1a90535dfcabb00e81081. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input format":"Box","Training strategy":"Training Methods"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.