# Grounded Multimodal Named Entity Recognition on Twitter-FMNERG Source: [Grounded Multimodal Named Entity Recognition on Twitter-FMNERG]() Collection index: [llms.txt]() Data updated: 2026-04-23T10:18:03.594Z Evaluation of Grounded Multimodal Named Entity Recognition performance on the Twitter-FMNERG dataset, reporting metrics including Avg. SR (%), GMNER, MNER, and EEG. 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 | Grounded Multimodal Named Entity Recognition | | Dataset (including any stated split) | Twitter-FMNERG | | Metric | Avg. SR (%) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [Avg. SR (%)](): Higher is better; unit: PERCENT. - [GMNER](): Higher is better; unit: SCALAR. - [MNER](): Higher is better; unit: SCALAR. - [EEG](): Higher is better; unit: SCALAR. ## Recorded results | Method | Avg. SR (%) | Result date | Report paper | | --- | --- | --- | --- | | VKEL | 100 | Not recorded | 69e9a7aa55a3c32af89f7147 | | Qwen2.5-VL-7B (MoRE)\* | 100 | Not recorded | 69e9a7aa55a3c32af89f7147 | | MAKAR | 100 | Not recorded | 69e9a7aa55a3c32af89f7147 | | SAKE | 68.8 | Not recorded | 69e9a7aa55a3c32af89f7147 | | GPT4o | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | Gemini-2.5 Pro | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | Qwen3-VL-Plus | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | UMT-OD-EVG | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | UMGF-OD-EVG | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | ITA-OD-EVG | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | MMT5 / BART-OD-EVG | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | H-Index | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | TIGER | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | MQSPN | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | Qwen2.5-VL-7B (SFT)\* | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | GEM | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | UnCo | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | | Qwen2.5-VL-7B (SFT+GRPO)\* | 0 | Not recorded | 69e9a7aa55a3c32af89f7147 | ### VKEL Avg. SR (%): 100. Result ID: 69e9f1c047f34ce924d1bb2c. [Reporting paper]() Recorded evaluation configuration: {"Category":"External Knowledge Exploration","E2E":false,"Using LLM":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B (MoRE)\* Avg. SR (%): 100. Result ID: 69e9f1c047f34ce924d1bb2d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"External Knowledge Exploration","E2E":false,"Using LLM":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### MAKAR Avg. SR (%): 100. Result ID: 69e9f1c047f34ce924d1bb31. [Reporting paper]() Recorded evaluation configuration: {"Category":"LLM Agent","E2E":false,"Using LLM":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### SAKE Avg. SR (%): 68.8. Result ID: 69e9f1c047f34ce924d1bb32. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"LLM Agent","E2E":true,"Using LLM":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### GPT4o Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb21. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Few-shot MLLM","E2E":false,"Using LLM":"GPT4o"}. This record does not assert independent reproduction. ### Gemini-2.5 Pro Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb22. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Few-shot MLLM","E2E":false,"Using LLM":"Gemini-2.5 Pro"}. This record does not assert independent reproduction. ### Qwen3-VL-Plus Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb23. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Few-shot MLLM","E2E":false,"Using LLM":"Qwen3-VL-Plus"}. This record does not assert independent reproduction. ### UMT-OD-EVG Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb24. [Reporting paper]() Recorded evaluation configuration: {"Category":"Supervised Fine-tuning","E2E":false}. This record does not assert independent reproduction. ### UMGF-OD-EVG Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb25. [Reporting paper]() Recorded evaluation configuration: {"Category":"Supervised Fine-tuning","E2E":false}. This record does not assert independent reproduction. ### ITA-OD-EVG Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb26. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Supervised Fine-tuning","E2E":false}. This record does not assert independent reproduction. ### MMT5 / BART-OD-EVG Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb27. [Reporting paper]() Recorded evaluation configuration: {"Category":"Supervised Fine-tuning","E2E":false}. This record does not assert independent reproduction. ### H-Index Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb28. [Reporting paper]() Recorded evaluation configuration: {"Category":"Supervised Fine-tuning","E2E":true}. This record does not assert independent reproduction. ### TIGER Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb29. [Reporting paper]() Recorded evaluation configuration: {"Category":"Supervised Fine-tuning","E2E":true}. This record does not assert independent reproduction. ### MQSPN Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb2a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Supervised Fine-tuning","E2E":true}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B (SFT)\* Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb2b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Supervised Fine-tuning","E2E":true,"Using LLM":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### GEM Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb2e. [Reporting paper]() Recorded evaluation configuration: {"Category":"Internal Knowledge Exploitation","E2E":false,"Using LLM":"ChatGPT"}. This record does not assert independent reproduction. ### UnCo Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb2f. [Reporting paper]() Recorded evaluation configuration: {"Category":"Internal Knowledge Exploitation","E2E":false,"Using LLM":"Gemini-2.5 Pro"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B (SFT+GRPO)\* Avg. SR (%): 0. Result ID: 69e9f1c047f34ce924d1bb30. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Internal Knowledge Exploitation","E2E":true,"Using LLM":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.