# Text-to-Image Generation on PIE-Bench Neg

Source: [Text-to-Image Generation on PIE-Bench Neg](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-09T06:38:40.604Z

Evaluation of Text-to-Image generation models on the PIE-Bench Neg dataset, utilizing metrics derived from positive prompt adherence, negative prompt adherence, VLM alignment, and aesthetic quality scores.

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 | Text-to-Image Generation |
| Dataset (including any stated split) | PIE-Bench Neg |
| Metric | Positive Prompt CLIP Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [Positive Prompt CLIP Score](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Positive%20Prompt%20CLIP%20Score>): Higher is better; unit: SCALAR.

- [Positive Prompt IR Score](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Positive%20Prompt%20IR%20Score>): Higher is better; unit: SCALAR.

- [Positive Prompt HPSv2 Score](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Positive%20Prompt%20HPSv2%20Score>): Higher is better; unit: SCALAR.

- [Positive Prompt VLM Alignment](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Positive%20Prompt%20VLM%20Alignment>): Higher is better; unit: SCALAR.

- [Negative Prompt CLIP Score](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Negative%20Prompt%20CLIP%20Score>): Higher is better; unit: SCALAR.

- [Negative Prompt IR Score](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Negative%20Prompt%20IR%20Score>): Lower is better; unit: SCALAR.

- [Negative Prompt HPSv2 Score](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Negative%20Prompt%20HPSv2%20Score>): Higher is better; unit: SCALAR.

- [Negative Prompt VLM Alignment](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Negative%20Prompt%20VLM%20Alignment>): Higher is better; unit: SCALAR.

- [Overall VLM Alignment](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Overall%20VLM%20Alignment>): Higher is better; unit: SCALAR.

- [Aesthetic Score](<https://www.sota2.com/research/sota/text-to-image-generation-on-pie-bench-neg.md?metric=Aesthetic%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Positive Prompt CLIP Score | Result date | Report paper |
| --- | --- | --- | --- |
| DNG | 0.2745 | Not recorded | 68fa75e31e209131cea6a7a0 |
| CCFG | 0.2739 | Not recorded | 68fa75e31e209131cea6a7a0 |
| Perp-Neg | 0.2735 | Not recorded | 68fa75e31e209131cea6a7a0 |
| SLD | 0.2733 | Not recorded | 68fa75e31e209131cea6a7a0 |
| SAFREE | 0.2729 | Not recorded | 68fa75e31e209131cea6a7a0 |
| Perp-Neg | 0.2724 | Not recorded | 68fa75e31e209131cea6a7a0 |
| CCFG | 0.2722 | Not recorded | 68fa75e31e209131cea6a7a0 |
| CCFG | 0.2703 | Not recorded | 68fa75e31e209131cea6a7a0 |
| DNG | 0.2702 | Not recorded | 68fa75e31e209131cea6a7a0 |
| SLD | 0.2687 | Not recorded | 68fa75e31e209131cea6a7a0 |
| NP | 0.2665 | Not recorded | 68fa75e31e209131cea6a7a0 |
| SAFREE | 0.2658 | Not recorded | 68fa75e31e209131cea6a7a0 |
| NP | 0.2616 | Not recorded | 68fa75e31e209131cea6a7a0 |
| NP | 0.2612 | Not recorded | 68fa75e31e209131cea6a7a0 |

### DNG

Positive Prompt CLIP Score: 0.2745. Result ID: 6a4f41e360b88bb339e41a84.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa795d1e209131cea7d46d.md>)

Recorded evaluation configuration: {"Backbone":"SDXL","stand-alone":"true"}. This record does not assert independent reproduction.

### CCFG

Positive Prompt CLIP Score: 0.2739. Result ID: 6a4f41e360b88bb339e41a88.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

Recorded evaluation configuration: {"Backbone":"SDXL","stand-alone":"true"}. This record does not assert independent reproduction.

### Perp-Neg

Positive Prompt CLIP Score: 0.2735. Result ID: 6a4f41e360b88bb339e41a85.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f231e209131cea43f80.md>)

Recorded evaluation configuration: {"Backbone":"SDXL","stand-alone":"false"}. This record does not assert independent reproduction.

### SLD

Positive Prompt CLIP Score: 0.2733. Result ID: 6a4f41e360b88bb339e41a86.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f231e209131cea43f5d.md>)

[Implementation](<https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion_safe>)

Recorded evaluation configuration: {"Backbone":"SDXL","stand-alone":"false"}. This record does not assert independent reproduction.

### SAFREE

Positive Prompt CLIP Score: 0.2729. Result ID: 6a4f41e360b88bb339e41a87.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77721e209131cea7326a.md>)

Recorded evaluation configuration: {"Backbone":"SDXL","stand-alone":"false"}. This record does not assert independent reproduction.

### Perp-Neg

Positive Prompt CLIP Score: 0.2724. Result ID: 6a4f41e360b88bb339e41a7f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f231e209131cea43f80.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 1.5","stand-alone":"false"}. This record does not assert independent reproduction.

### CCFG

Positive Prompt CLIP Score: 0.2722. Result ID: 6a4f41e360b88bb339e41a9b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

Recorded evaluation configuration: {"Base Model":"StableDiffusion 3"}. This record does not assert independent reproduction.

### CCFG

Positive Prompt CLIP Score: 0.2703. Result ID: 6a4f41e360b88bb339e41a82.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 1.5","stand-alone":"true"}. This record does not assert independent reproduction.

### DNG

Positive Prompt CLIP Score: 0.2702. Result ID: 6a4f41e360b88bb339e41a7e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa795d1e209131cea7d46d.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 1.5","stand-alone":"true"}. This record does not assert independent reproduction.

### SLD

Positive Prompt CLIP Score: 0.2687. Result ID: 6a4f41e360b88bb339e41a80.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f231e209131cea43f5d.md>)

[Implementation](<https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion_safe>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 1.5","stand-alone":"false"}. This record does not assert independent reproduction.

### NP

Positive Prompt CLIP Score: 0.2665. Result ID: 6a4f41e360b88bb339e41a83.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fc61e209131cea47556.md>)

Recorded evaluation configuration: {"Backbone":"SDXL","stand-alone":"true"}. This record does not assert independent reproduction.

### SAFREE

Positive Prompt CLIP Score: 0.2658. Result ID: 6a4f41e360b88bb339e41a81.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77721e209131cea7326a.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 1.5","stand-alone":"false"}. This record does not assert independent reproduction.

### NP

Positive Prompt CLIP Score: 0.2616. Result ID: 6a4f41e360b88bb339e41a9a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fc61e209131cea47556.md>)

Recorded evaluation configuration: {"Base Model":"StableDiffusion 3"}. This record does not assert independent reproduction.

### NP

Positive Prompt CLIP Score: 0.2612. Result ID: 6a4f41e360b88bb339e41a7d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e31e209131cea6a7a0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fc61e209131cea47556.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 1.5","stand-alone":"true"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.