# Single Product shopping assistance on SHOPPING COMPANION benchmark

Source: [Single Product shopping assistance on SHOPPING COMPANION benchmark](<https://www.sota2.com/research/sota/single-product-shopping-assistance-on-shopping-companion-benchmark>)

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

Data updated: 2026-03-17T11:15:17.724Z

Evaluation of performance for single product shopping assistance tasks using the SHOPPING COMPANION benchmark, measured by Accuracy and Success Rate.

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 | Single Product shopping assistance |
| Dataset (including any stated split) | SHOPPING COMPANION benchmark |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/single-product-shopping-assistance-on-shopping-companion-benchmark.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Success Rate](<https://www.sota2.com/research/sota/single-product-shopping-assistance-on-shopping-companion-benchmark.md?metric=Success%20Rate>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen3-4B-LoRA + RL (Dual&Tool-wise Reward) | 90 | Not recorded | 69b8e0590636e5bf4afbef4c |
| Qwen3-4B-LoRA + RL (Dual-reward) | 89 | Not recorded | 69b8e0590636e5bf4afbef4c |
| GPT-4.1 | 88 | Not recorded | 69b8e0590636e5bf4afbef4c |
| GPT-5 | 82 | Not recorded | 69b8e0590636e5bf4afbef4c |
| Qwen3-4B-LoRA | 82 | Not recorded | 69b8e0590636e5bf4afbef4c |
| Qwen3-Max | 80 | Not recorded | 69b8e0590636e5bf4afbef4c |
| GPT-4o | 79 | Not recorded | 69b8e0590636e5bf4afbef4c |
| Qwen3-Next-80B-A3B | 63 | Not recorded | 69b8e0590636e5bf4afbef4c |
| Qwen3-30B-A3B | 60 | Not recorded | 69b8e0590636e5bf4afbef4c |
| Qwen3-4B | 49 | Not recorded | 69b8e0590636e5bf4afbef4c |

### Qwen3-4B-LoRA + RL (Dual&Tool-wise Reward)

Accuracy: 90. Result ID: 69b937b047f34ce924a2d820.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Ours"}. This record does not assert independent reproduction.

### Qwen3-4B-LoRA + RL (Dual-reward)

Accuracy: 89. Result ID: 69b937b047f34ce924a2d81f.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Ours"}. This record does not assert independent reproduction.

### GPT-4.1

Accuracy: 88. Result ID: 69b937b047f34ce924a2d818.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Closed"}. This record does not assert independent reproduction.

### GPT-5

Accuracy: 82. Result ID: 69b937b047f34ce924a2d817.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Closed"}. This record does not assert independent reproduction.

### Qwen3-4B-LoRA

Accuracy: 82. Result ID: 69b937b047f34ce924a2d81e.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Ours"}. This record does not assert independent reproduction.

### Qwen3-Max

Accuracy: 80. Result ID: 69b937b047f34ce924a2d81a.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Closed"}. This record does not assert independent reproduction.

### GPT-4o

Accuracy: 79. Result ID: 69b937b047f34ce924a2d819.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Closed"}. This record does not assert independent reproduction.

### Qwen3-Next-80B-A3B

Accuracy: 63. Result ID: 69b937b047f34ce924a2d81b.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Open"}. This record does not assert independent reproduction.

### Qwen3-30B-A3B

Accuracy: 60. Result ID: 69b937b047f34ce924a2d81c.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Open"}. This record does not assert independent reproduction.

### Qwen3-4B

Accuracy: 49. Result ID: 69b937b047f34ce924a2d81d.

[Reporting paper](<https://www.sota2.com/research/paper/69b8e0590636e5bf4afbef4c.md>)

Recorded evaluation configuration: {"Category":"Open"}. This record does not assert independent reproduction.

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