# Hateful Meme Identification on Extended BHM (test) Source: [Hateful Meme Identification on Extended BHM (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:58:45.631Z Evaluation of model performance on the test split of the Extended BHM dataset for Hateful Meme Identification, measured using Precision, Recall, and F1 score. 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 | Hateful Meme Identification | | Dataset (including any stated split) | Extended BHM (test) | | Metric | Precision | | 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 - [Precision](): Higher is better; unit: PERCENT. - [Recall](): Higher is better; unit: PERCENT. - [F1 Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Precision | Result date | Report paper | | --- | --- | --- | --- | | CLIP + XLM-R-L | 79 | Not recorded | 699d307c6017c4460ffab3bb | | RAG-Fused DORA | 79 | Not recorded | 699d307c6017c4460ffab3bb | | CLIP + XGLM | 78 | Not recorded | 699d307c6017c4460ffab3bb | | FAISS-Based k-NN | 78 | Not recorded | 699d307c6017c4460ffab3bb | | MAF | 77 | Not recorded | 699d307c6017c4460ffab3bb | | DINOv2 + XLM-R | 77 | Not recorded | 699d307c6017c4460ffab3bb | | DORA | 76 | Not recorded | 699d307c6017c4460ffab3bb | | DINOv2 + XGLM | 76 | Not recorded | 699d307c6017c4460ffab3bb | | XGLM-564M | 75 | Not recorded | 699d307c6017c4460ffab3bb | | mDeBERTa-v3 | 74 | Not recorded | 699d307c6017c4460ffab3bb | | XLM-R-L | 73 | Not recorded | 699d307c6017c4460ffab3bb | | DINOv2-Base | 65 | Not recorded | 699d307c6017c4460ffab3bb | | ConvNeXT-Base | 63 | Not recorded | 699d307c6017c4460ffab3bb | | LLaVA | 55 | Not recorded | 699d307c6017c4460ffab3bb | | RAG-Prompted LLaVA | 55 | Not recorded | 699d307c6017c4460ffab3bb | | LLaVA | 53 | Not recorded | 699d307c6017c4460ffab3bb | | CLIP (ViT-B/32) | 50 | Not recorded | 699d307c6017c4460ffab3bb | ### CLIP + XLM-R-L Precision: 79. Result ID: 699e3b735698665ec4f83861. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"xDORA","Category":"Proposed Supervised Framework Variants"}. This record does not assert independent reproduction. ### RAG-Fused DORA Precision: 79. Result ID: 699e3b735698665ec4f83865. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Proposed Retrieval-Based Methods"}. This record does not assert independent reproduction. ### CLIP + XGLM Precision: 78. Result ID: 699e3b735698665ec4f83860. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"xDORA","Category":"Proposed Supervised Framework Variants"}. This record does not assert independent reproduction. ### FAISS-Based k-NN Precision: 78. Result ID: 699e3b735698665ec4f83864. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Proposed Retrieval-Based Methods"}. This record does not assert independent reproduction. ### MAF Precision: 77. Result ID: 699e3b735698665ec4f8385d. [Reporting paper]() Recorded evaluation configuration: {"Category":"Multimodal"}. This record does not assert independent reproduction. ### DINOv2 + XLM-R Precision: 77. Result ID: 699e3b735698665ec4f83863. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"xDORA","Category":"Proposed Supervised Framework Variants"}. This record does not assert independent reproduction. ### DORA Precision: 76. Result ID: 699e3b735698665ec4f8385c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Multimodal"}. This record does not assert independent reproduction. ### DINOv2 + XGLM Precision: 76. Result ID: 699e3b735698665ec4f83862. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Framework":"xDORA","Category":"Proposed Supervised Framework Variants"}. This record does not assert independent reproduction. ### XGLM-564M Precision: 75. Result ID: 699e3b735698665ec4f83859. [Reporting paper]() Recorded evaluation configuration: {"Category":"Text Only"}. This record does not assert independent reproduction. ### mDeBERTa-v3 Precision: 74. Result ID: 699e3b735698665ec4f8385b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Text Only"}. This record does not assert independent reproduction. ### XLM-R-L Precision: 73. Result ID: 699e3b735698665ec4f8385a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Text Only"}. This record does not assert independent reproduction. ### DINOv2-Base Precision: 65. Result ID: 699e3b735698665ec4f83857. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Vision Only"}. This record does not assert independent reproduction. ### ConvNeXT-Base Precision: 63. Result ID: 699e3b735698665ec4f83858. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Vision Only"}. This record does not assert independent reproduction. ### LLaVA Precision: 55. Result ID: 699e3b735698665ec4f8385e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"LVLM","Zero-shot":true}. This record does not assert independent reproduction. ### RAG-Prompted LLaVA Precision: 55. Result ID: 699e3b735698665ec4f83866. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Proposed Retrieval-Based Methods"}. This record does not assert independent reproduction. ### LLaVA Precision: 53. Result ID: 699e3b735698665ec4f8385f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"LVLM","Few-shot":true}. This record does not assert independent reproduction. ### CLIP (ViT-B/32) Precision: 50. Result ID: 699e3b735698665ec4f83856. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Vision Only"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.