# Interaction Recognition on ImageNet-1k 1 object

Source: [Interaction Recognition on ImageNet-1k 1 object](<https://www.sota2.com/research/sota/interaction-recognition-on-imagenet-1k-1-object>)

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

Data updated: 2026-05-08T10:48:12.616Z

Evaluation of Interaction Recognition performance on the ImageNet-1k 1 object subset/variant.

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 | Interaction Recognition |
| Dataset (including any stated split) | ImageNet-1k 1 object |
| Metric | Interaction Recognition |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 26 of 26 |
| 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

- [Interaction Recognition](<https://www.sota2.com/research/sota/interaction-recognition-on-imagenet-1k-1-object.md?metric=Interaction%20Recognition>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Interaction Recognition | Result date | Report paper |
| --- | --- | --- | --- |
| MaskSigLIP | 89 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 89 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 88 | Not recorded | 69fd6e215497b3b670c894d8 |
| FIxLIP | 87 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 86 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 85 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 84 | Not recorded | 69fd6e215497b3b670c894d8 |
| Attention | 84 | Not recorded | 69fd6e215497b3b670c894d8 |
| FIxLIP | 80 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 79 | Not recorded | 69fd6e215497b3b670c894d8 |
| Attention | 77 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 76 | Not recorded | 69fd6e215497b3b670c894d8 |
| FIxLIP | 76 | Not recorded | 69fd6e215497b3b670c894d8 |
| Grad-ECLIP | 76 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 75 | Not recorded | 69fd6e215497b3b670c894d8 |
| Grad-ECLIP | 71 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 71 | Not recorded | 69fd6e215497b3b670c894d8 |
| FIxLIP | 70 | Not recorded | 69fd6e215497b3b670c894d8 |
| Grad-ESigLIP | 68 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 60 | Not recorded | 69fd6e215497b3b670c894d8 |
| METAGAME | 57 | Not recorded | 69fd6e215497b3b670c894d8 |
| Grad-ESigLIP | 54 | Not recorded | 69fd6e215497b3b670c894d8 |
| MaskSigLIP | 53 | Not recorded | 69fd6e215497b3b670c894d8 |
| Attention | 50 | Not recorded | 69fd6e215497b3b670c894d8 |
| Attention | 47 | Not recorded | 69fd6e215497b3b670c894d8 |
| MaskCLIP | 30 | Not recorded | 69fd6e215497b3b670c894d8 |

### MaskSigLIP

Interaction Recognition: 89. Result ID: 69fdbef047f34ce924e172b4.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-L/16)"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 89. Result ID: 69fdbef047f34ce924e172bc.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-B/32)","Base Method":"MaskSigLIP"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 88. Result ID: 69fdbef047f34ce924e172b5.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-L/16)","Base Method":"MaskSigLIP"}. This record does not assert independent reproduction.

### FIxLIP

Interaction Recognition: 87. Result ID: 69fdbef047f34ce924e172bf.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a6e1e209131cea832b5.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-B/32)"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 86. Result ID: 69fdbef047f34ce924e17285.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"CLIP (ViT-B/16)","Explanation Basis":"MaskCLIP"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 85. Result ID: 69fdbef047f34ce924e1728a.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"MetaCLIP-2 (ViT-H/14)","Explanation Basis":"Attention"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 84. Result ID: 69fdbef047f34ce924e17283.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"CLIP (ViT-B/16)","Explanation Basis":"Attention"}. This record does not assert independent reproduction.

### Attention

Interaction Recognition: 84. Result ID: 69fdbef047f34ce924e17289.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69a11e209131cea233a8.md>)

Recorded evaluation configuration: {"Model (Size)":"MetaCLIP-2 (ViT-H/14)"}. This record does not assert independent reproduction.

### FIxLIP

Interaction Recognition: 80. Result ID: 69fdbef047f34ce924e172b8.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a6e1e209131cea832b5.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-L/16)"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 79. Result ID: 69fdbef047f34ce924e1728c.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"MetaCLIP-2 (ViT-H/14)","Explanation Basis":"Grad-ECLIP"}. This record does not assert independent reproduction.

### Attention

Interaction Recognition: 77. Result ID: 69fdbef047f34ce924e17282.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69a11e209131cea233a8.md>)

Recorded evaluation configuration: {"Model (Size)":"CLIP (ViT-B/16)"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 76. Result ID: 69fdbef047f34ce924e17287.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"CLIP (ViT-B/16)","Explanation Basis":"Grad-ECLIP"}. This record does not assert independent reproduction.

### FIxLIP

Interaction Recognition: 76. Result ID: 69fdbef047f34ce924e17288.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a6e1e209131cea832b5.md>)

Recorded evaluation configuration: {"Model (Size)":"CLIP (ViT-B/16)"}. This record does not assert independent reproduction.

### Grad-ECLIP

Interaction Recognition: 76. Result ID: 69fdbef047f34ce924e1728b.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"MetaCLIP-2 (ViT-H/14)"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 75. Result ID: 69fdbef047f34ce924e172b7.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-L/16)","Base Method":"Grad-ESigLIP"}. This record does not assert independent reproduction.

### Grad-ECLIP

Interaction Recognition: 71. Result ID: 69fdbef047f34ce924e17286.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"CLIP (ViT-B/16)"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 71. Result ID: 69fdbef047f34ce924e172be.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-B/32)","Base Method":"Grad-ESigLIP"}. This record does not assert independent reproduction.

### FIxLIP

Interaction Recognition: 70. Result ID: 69fdbef047f34ce924e1728d.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a6e1e209131cea832b5.md>)

Recorded evaluation configuration: {"Model (Size)":"MetaCLIP-2 (ViT-H/14)"}. This record does not assert independent reproduction.

### Grad-ESigLIP

Interaction Recognition: 68. Result ID: 69fdbef047f34ce924e172bd.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-B/32)"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 60. Result ID: 69fdbef047f34ce924e172b3.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-L/16)","Base Method":"Attention"}. This record does not assert independent reproduction.

### METAGAME

Interaction Recognition: 57. Result ID: 69fdbef047f34ce924e172ba.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-B/32)","Base Method":"Attention"}. This record does not assert independent reproduction.

### Grad-ESigLIP

Interaction Recognition: 54. Result ID: 69fdbef047f34ce924e172b6.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-L/16)"}. This record does not assert independent reproduction.

### MaskSigLIP

Interaction Recognition: 53. Result ID: 69fdbef047f34ce924e172bb.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-B/32)"}. This record does not assert independent reproduction.

### Attention

Interaction Recognition: 50. Result ID: 69fdbef047f34ce924e172b9.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69a11e209131cea233a8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-B/32)"}. This record does not assert independent reproduction.

### Attention

Interaction Recognition: 47. Result ID: 69fdbef047f34ce924e172b2.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69a11e209131cea233a8.md>)

Recorded evaluation configuration: {"Model (Size)":"SigLIP-2 (ViT-L/16)"}. This record does not assert independent reproduction.

### MaskCLIP

Interaction Recognition: 30. Result ID: 69fdbef047f34ce924e17284.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e215497b3b670c894d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f0a1e209131cea437fe.md>)

[Implementation](<https://github.com/LightDXY/MaskCLIP>)

Recorded evaluation configuration: {"Model (Size)":"CLIP (ViT-B/16)"}. This record does not assert independent reproduction.

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