# Detailed Image Captioning on MS COCO 2014 (val) Source: [Detailed Image Captioning on MS COCO 2014 (val)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:38:19.969Z Evaluation of Detailed Image Captioning performance on the validation split of the MS COCO 2014 dataset, assessed using CHAIRi, CHAIRs, and Cover metrics. 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 | Detailed Image Captioning | | Dataset (including any stated split) | MS COCO 2014 (val) | | Metric | CHAIRi | | Unit | SCALAR | | 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 - [CHAIRi](): Higher is better; unit: SCALAR. - [CHAIRs](): Higher is better; unit: SCALAR. - [Cover](): Higher is better; unit: PERCENT. ## Recorded results | Method | CHAIRi | Result date | Report paper | | --- | --- | --- | --- | | LLaVA | 10.1 | Not recorded | 68fa729b1e209131cea5779f | | LLaVA Contrastive | 10.1 | Not recorded | 68fa729b1e209131cea5779f | | LLaVA PMI | 9.8 | Not recorded | 68fa729b1e209131cea5779f | | LLaVA M3ID | 8.8 | Not recorded | 68fa729b1e209131cea5779f | | LLaVA PMI | 8.6 | Not recorded | 68fa729b1e209131cea5779f | | LLaVA M3ID + DPO | 8.5 | Not recorded | 68fa729b1e209131cea5779f | | LLaVA | 8.5 | Not recorded | 68fa729b1e209131cea5779f | | LLaVA Contrastive | 8.1 | Not recorded | 68fa729b1e209131cea5779f | | LLaVA M3ID | 7.8 | Not recorded | 68fa729b1e209131cea5779f | | LLaVA M3ID + DPO | 7.5 | Not recorded | 68fa729b1e209131cea5779f | ### LLaVA CHAIRi: 10.1. Result ID: 699262b03a6c3148490e953b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-7B","Decoding Strategy":"Standard"}. This record does not assert independent reproduction. ### LLaVA Contrastive CHAIRi: 10.1. Result ID: 699262b03a6c3148490e953d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-7B","Decoding Strategy":"Contrastive"}. This record does not assert independent reproduction. ### LLaVA PMI CHAIRi: 9.8. Result ID: 699262b03a6c3148490e953c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-7B","Decoding Strategy":"PMI"}. This record does not assert independent reproduction. ### LLaVA M3ID CHAIRi: 8.8. Result ID: 699262b03a6c3148490e953e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-7B","Decoding Strategy":"M3ID"}. This record does not assert independent reproduction. ### LLaVA PMI CHAIRi: 8.6. Result ID: 699262b03a6c3148490e9541. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-13B","Decoding Strategy":"PMI"}. This record does not assert independent reproduction. ### LLaVA M3ID + DPO CHAIRi: 8.5. Result ID: 699262b03a6c3148490e953f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-7B","Decoding Strategy":"M3ID + DPO"}. This record does not assert independent reproduction. ### LLaVA CHAIRi: 8.5. Result ID: 699262b03a6c3148490e9540. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-13B","Decoding Strategy":"Standard"}. This record does not assert independent reproduction. ### LLaVA Contrastive CHAIRi: 8.1. Result ID: 699262b03a6c3148490e9542. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-13B","Decoding Strategy":"Contrastive"}. This record does not assert independent reproduction. ### LLaVA M3ID CHAIRi: 7.8. Result ID: 699262b03a6c3148490e9543. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-13B","Decoding Strategy":"M3ID"}. This record does not assert independent reproduction. ### LLaVA M3ID + DPO CHAIRi: 7.5. Result ID: 699262b03a6c3148490e9544. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-13B","Decoding Strategy":"M3ID + DPO"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.