# element-level text-to-image alignment evaluation on GenAI-Bench Source: [element-level text-to-image alignment evaluation on GenAI-Bench]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:03:14.102Z Evaluation of element-level alignment between text prompts and generated images using the GenAI-Bench dataset, measured by correlation (SRCC, PLCC) and accuracy 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 | element-level text-to-image alignment evaluation | | Dataset (including any stated split) | GenAI-Bench | | Metric | SRCC | | Unit | SCALAR | | 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 - [SRCC](): Higher is better; unit: SCALAR. - [PLCC](): Higher is better; unit: SCALAR. - [Acc](): Higher is better; unit: PERCENT. ## Recorded results | Method | SRCC | Result date | Report paper | | --- | --- | --- | --- | | REVEALER | 0.749 | Not recorded | 69535c9253fdd5c36173da22 | | REVEALER | 0.734 | Not recorded | 69535c9253fdd5c36173da22 | | REVEALER | 0.723 | Not recorded | 69535c9253fdd5c36173da22 | | TIFA | 0.711 | Not recorded | 69535c9253fdd5c36173da22 | | VQAScore | 0.708 | Not recorded | 69535c9253fdd5c36173da22 | | VQ2 | 0.704 | Not recorded | 69535c9253fdd5c36173da22 | | VQ2 | 0.702 | Not recorded | 69535c9253fdd5c36173da22 | | REVEALER | 0.696 | Not recorded | 69535c9253fdd5c36173da22 | | TIFA | 0.692 | Not recorded | 69535c9253fdd5c36173da22 | | VIEScore | 0.692 | Not recorded | 69535c9253fdd5c36173da22 | | SFT | 0.678 | Not recorded | 69535c9253fdd5c36173da22 | | SFT | 0.671 | Not recorded | 69535c9253fdd5c36173da22 | | REVEALER | 0.668 | Not recorded | 69535c9253fdd5c36173da22 | | FGA-BLIP2 | 0.653 | Not recorded | 69535c9253fdd5c36173da22 | | SFT | 0.629 | Not recorded | 69535c9253fdd5c36173da22 | | TIFA | 0.607 | Not recorded | 69535c9253fdd5c36173da22 | | VQ2 | 0.349 | Not recorded | 69535c9253fdd5c36173da22 | ### REVEALER SRCC: 0.749. Result ID: 699802865698665ec45c0693. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B-Instruct"}. This record does not assert independent reproduction. ### REVEALER SRCC: 0.734. Result ID: 699802865698665ec45c0691. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"InternVL3-8B-hf"}. This record does not assert independent reproduction. ### REVEALER SRCC: 0.723. Result ID: 699802865698665ec45c068f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DINO, Gemini2.5-pro"}. This record does not assert independent reproduction. ### TIFA SRCC: 0.711. Result ID: 699802865698665ec45c0683. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Gemini2.5-pro","Paradigm":"Zero-Shot"}. This record does not assert independent reproduction. ### VQAScore SRCC: 0.708. Result ID: 699802865698665ec45c0689. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"CLIP-FlanT5-XXL","Paradigm":"Zero-Shot"}. This record does not assert independent reproduction. ### VQ2 SRCC: 0.704. Result ID: 699802865698665ec45c0687. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-72B","Paradigm":"Zero-Shot"}. This record does not assert independent reproduction. ### VQ2 SRCC: 0.702. Result ID: 699802865698665ec45c0686. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-4o","Paradigm":"Zero-Shot"}. This record does not assert independent reproduction. ### REVEALER SRCC: 0.696. Result ID: 699802865698665ec45c0692. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-4B-Instruct"}. This record does not assert independent reproduction. ### TIFA SRCC: 0.692. Result ID: 699802865698665ec45c0684. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-72B","Paradigm":"Zero-Shot"}. This record does not assert independent reproduction. ### VIEScore SRCC: 0.692. Result ID: 699802865698665ec45c068a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-4o","Paradigm":"Zero-Shot"}. This record does not assert independent reproduction. ### SFT SRCC: 0.678. Result ID: 699802865698665ec45c068d. [Reporting paper]() Recorded evaluation configuration: {"Model":"InternVL3-8B-hf","Paradigm":"Supervised Fine-Tuning"}. This record does not assert independent reproduction. ### SFT SRCC: 0.671. Result ID: 699802865698665ec45c068c. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B-Instruct","Paradigm":"Supervised Fine-Tuning"}. This record does not assert independent reproduction. ### REVEALER SRCC: 0.668. Result ID: 699802865698665ec45c0690. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"InternVL3-2B-hf"}. This record does not assert independent reproduction. ### FGA-BLIP2 SRCC: 0.653. Result ID: 699802865698665ec45c068b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"BLIP2","Paradigm":"Supervised Fine-Tuning"}. This record does not assert independent reproduction. ### SFT SRCC: 0.629. Result ID: 699802865698665ec45c068e. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaVA-v1.6-7B-hf","Paradigm":"Supervised Fine-Tuning"}. This record does not assert independent reproduction. ### TIFA SRCC: 0.607. Result ID: 699802865698665ec45c0685. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"mPLUG-Owl3","Paradigm":"Zero-Shot"}. This record does not assert independent reproduction. ### VQ2 SRCC: 0.349. Result ID: 699802865698665ec45c0688. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaVA-v1.6-7B-hf","Paradigm":"Zero-Shot"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.