# Vision-Language Probing on VL-CheckList (test)

Source: [Vision-Language Probing on VL-CheckList (test)](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test>)

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

Data updated: 2026-02-26T05:54:09.575Z

Evaluation of model probing capabilities across various linguistic and visual attributes, objects, and relations using the test split of the VL-CheckList dataset.

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 | Vision-Language Probing |
| Dataset (including any stated split) | VL-CheckList (test) |
| Metric | Object: Avg |
| 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

- [Object: Avg](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Object%3A%20Avg>): Higher is better; unit: PERCENT.

- [Attribute: Action](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Attribute%3A%20Action>): Higher is better; unit: PERCENT.

- [Attribute: Color](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Attribute%3A%20Color>): Higher is better; unit: PERCENT.

- [Attribute: Material](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Attribute%3A%20Material>): Higher is better; unit: PERCENT.

- [Attribute: Size](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Attribute%3A%20Size>): Higher is better; unit: PERCENT.

- [Attribute: State](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Attribute%3A%20State>): Higher is better; unit: PERCENT.

- [Attribute: Avg](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Attribute%3A%20Avg>): Higher is better; unit: PERCENT.

- [Object: Location](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Object%3A%20Location>): Higher is better; unit: PERCENT.

- [Object: Size](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Object%3A%20Size>): Higher is better; unit: PERCENT.

- [Relation: Action](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Relation%3A%20Action>): Higher is better; unit: PERCENT.

- [Relation: Spatial](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Relation%3A%20Spatial>): Higher is better; unit: PERCENT.

- [Relation: Avg](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Relation%3A%20Avg>): Higher is better; unit: PERCENT.

- [Total Avg](<https://www.sota2.com/research/sota/vision-language-probing-on-vl-checklist-test.md?metric=Total%20Avg>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Object: Avg | Result date | Report paper |
| --- | --- | --- | --- |
| CE-XVLM | 86.9 | Not recorded | 68fa730b1e209131cea5a166 |
| CE-CLIP+ | 86.3 | Not recorded | 68fa730b1e209131cea5a166 |
| XVLM-coco | 85.8 | Not recorded | 68fa730b1e209131cea5a166 |
| CLIP-SVLC | 85 | Not recorded | 68fa730b1e209131cea5a166 |
| BLIP2 | 84.9 | Not recorded | 68fa730b1e209131cea5a166 |
| CE-CLIP | 84.6 | Not recorded | 68fa730b1e209131cea5a166 |
| BEIT3 | 84.5 | Not recorded | 68fa730b1e209131cea5a166 |
| MiniGPT-4 | 84.2 | Not recorded | 68fa730b1e209131cea5a166 |
| NegCLIP | 84.1 | Not recorded | 68fa730b1e209131cea5a166 |
| CLIP-SGVL | 82.6 | Not recorded | 68fa730b1e209131cea5a166 |
| BLIP | 82.2 | Not recorded | 68fa730b1e209131cea5a166 |
| UNITER | 81.9 | Not recorded | 68fa730b1e209131cea5a166 |
| ALBEF | 81.1 | Not recorded | 68fa730b1e209131cea5a166 |
| CLIP | 80 | Not recorded | 68fa730b1e209131cea5a166 |
| DAC-LLM | 77.3 | Not recorded | 68fa730b1e209131cea5a166 |
| DAC-SAM | 75.8 | Not recorded | 68fa730b1e209131cea5a166 |
| syn-CLIP | 69.4 | Not recorded | 68fa730b1e209131cea5a166 |

### CE-XVLM

Object: Avg: 86.9. Result ID: 6992bcd23a6c3148491122fe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

Recorded evaluation configuration: {"#Params":"216M"}. This record does not assert independent reproduction.

### CE-CLIP+

Object: Avg: 86.3. Result ID: 6992bcd23a6c314849112303.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Implementation](<https://github.com/lezhang7/Enhance-FineGrained>)

Recorded evaluation configuration: {"#Params":"151M"}. This record does not assert independent reproduction.

### XVLM-coco

Object: Avg: 85.8. Result ID: 6992bcd23a6c3148491122fd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea358ec.md>)

[Implementation](<https://github.com/zengyan-97/X-VLM>)

Recorded evaluation configuration: {"#Params":"216M"}. This record does not assert independent reproduction.

### CLIP-SVLC

Object: Avg: 85. Result ID: 6992bcd23a6c314849112300.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f961e209131cea46773.md>)

Recorded evaluation configuration: {"#Params":"151M"}. This record does not assert independent reproduction.

### BLIP2

Object: Avg: 84.9. Result ID: 6992bcd23a6c3148491122f7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea46fe6.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/blip2>)

Recorded evaluation configuration: {"#Params":"3.4B"}. This record does not assert independent reproduction.

### CE-CLIP

Object: Avg: 84.6. Result ID: 6992bcd23a6c314849112302.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Implementation](<https://github.com/lezhang7/Enhance-FineGrained>)

Recorded evaluation configuration: {"#Params":"151M"}. This record does not assert independent reproduction.

### BEIT3

Object: Avg: 84.5. Result ID: 6992bcd23a6c3148491122f6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d541e209131cea39310.md>)

[Implementation](<https://aka.ms/beit-3>)

Recorded evaluation configuration: {"#Params":"1.9B"}. This record does not assert independent reproduction.

### MiniGPT-4

Object: Avg: 84.2. Result ID: 6992bcd23a6c3148491122f8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea46fe6.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/blip2>)

Recorded evaluation configuration: {"#Params":"\>9B"}. This record does not assert independent reproduction.

### NegCLIP

Object: Avg: 84.1. Result ID: 6992bcd23a6c314849112301.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a7d.md>)

Recorded evaluation configuration: {"#Params":"151M"}. This record does not assert independent reproduction.

### CLIP-SGVL

Object: Avg: 82.6. Result ID: 6992bcd23a6c3148491122f9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f11e209131cea4e384.md>)

[Implementation](<https://alonmendelson.github.io/SGVL/>)

Recorded evaluation configuration: {"#Params":"\>151M"}. This record does not assert independent reproduction.

### BLIP

Object: Avg: 82.2. Result ID: 6992bcd23a6c3148491122f5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bd81e209131cea30874.md>)

[Implementation](<https://github.com/salesforce/BLIP>)

Recorded evaluation configuration: {"#Params":"583M"}. This record does not assert independent reproduction.

### UNITER

Object: Avg: 81.9. Result ID: 6992bcd23a6c3148491122f4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67f91e209131cea1989f.md>)

[Implementation](<https://github.com/ChenRocks/UNITER>)

Recorded evaluation configuration: {"#Params":"300M"}. This record does not assert independent reproduction.

### ALBEF

Object: Avg: 81.1. Result ID: 6992bcd23a6c3148491122f3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6aef1e209131cea2b2da.md>)

[Implementation](<https://github.com/salesforce/ALBEF>)

Recorded evaluation configuration: {"#Params":"210M"}. This record does not assert independent reproduction.

### CLIP

Object: Avg: 80. Result ID: 6992bcd23a6c3148491122ff.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"#Params":"151M"}. This record does not assert independent reproduction.

### DAC-LLM

Object: Avg: 77.3. Result ID: 6992bcd23a6c3148491122fb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f8c1e209131cea46253.md>)

Recorded evaluation configuration: {"#Params":"151M"}. This record does not assert independent reproduction.

### DAC-SAM

Object: Avg: 75.8. Result ID: 6992bcd23a6c3148491122fc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f8c1e209131cea46253.md>)

Recorded evaluation configuration: {"#Params":"151M"}. This record does not assert independent reproduction.

### syn-CLIP

Object: Avg: 69.4. Result ID: 6992bcd23a6c3148491122fa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa730b1e209131cea5a166.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa706f1e209131cea4af53.md>)

[Implementation](<https://github.com/uvavision/SyViC>)

Recorded evaluation configuration: {"#Params":"151M"}. This record does not assert independent reproduction.

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