# Web navigation on WebShop (50 instructions)

Source: [Web navigation on WebShop (50 instructions)](<https://www.sota2.com/research/sota/web-navigation-on-webshop-50-instructions>)

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

Data updated: 2026-02-26T06:03:45.807Z

Evaluation of web navigation performance on the 50-instruction subset of the WebShop dataset, measured by Score and Success Rate (SR).

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 | Web navigation |
| Dataset (including any stated split) | WebShop (50 instructions) |
| Metric | Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Score](<https://www.sota2.com/research/sota/web-navigation-on-webshop-50-instructions.md?metric=Score>): Higher is better; unit: PERCENT.

- [SR](<https://www.sota2.com/research/sota/web-navigation-on-webshop-50-instructions.md?metric=SR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Score | Result date | Report paper |
| --- | --- | --- | --- |
| Expert | 82.1 | Not recorded | 68fa738a1e209131cea5d15a |
| LATS (ReAct) | 75.9 | Not recorded | 68fa738a1e209131cea5d15a |
| Fine-tuning | 67.5 | Not recorded | 68fa738a1e209131cea5d15a |
| Reflexion | 64.2 | Not recorded | 68fa738a1e209131cea5d15a |
| IL+RL | 62.4 | Not recorded | 68fa738a1e209131cea5d15a |
| IL | 59.9 | Not recorded | 68fa738a1e209131cea5d15a |
| ReAct (best of k) | 59.1 | Not recorded | 68fa738a1e209131cea5d15a |
| ReAct | 53.8 | Not recorded | 68fa738a1e209131cea5d15a |

### Expert

Score: 82.1. Result ID: 6992f75e3a6c31484912bac5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d15a.md>)

Recorded evaluation configuration: {"type":"Human"}. This record does not assert independent reproduction.

### LATS (ReAct)

Score: 75.9. Result ID: 6992f75e3a6c31484912bac1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d15a.md>)

[Implementation](<https://github.com/lapisrocks/LanguageAgentTreeSearch>)

Recorded evaluation configuration: {"k":30,"n":5}. This record does not assert independent reproduction.

### Fine-tuning

Score: 67.5. Result ID: 6992f75e3a6c31484912bac4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72581e209131cea55cd7.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Reflexion

Score: 64.2. Result ID: 6992f75e3a6c31484912bac0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70cc1e209131cea4d38e.md>)

[Implementation](<https://github.com/noahshinn024/reflexion>)

Recorded evaluation configuration: {"k":30}. This record does not assert independent reproduction.

### IL+RL

Score: 62.4. Result ID: 6992f75e3a6c31484912bac3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e7b1e209131cea40279.md>)

Recorded evaluation configuration: {"type":"Reinforcement Learning"}. This record does not assert independent reproduction.

### IL

Score: 59.9. Result ID: 6992f75e3a6c31484912bac2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e7b1e209131cea40279.md>)

Recorded evaluation configuration: {"type":"Imitation Learning"}. This record does not assert independent reproduction.

### ReAct (best of k)

Score: 59.1. Result ID: 6992f75e3a6c31484912babf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"k":30}. This record does not assert independent reproduction.

### ReAct

Score: 53.8. Result ID: 6992f75e3a6c31484912babe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"trajectories":1}. This record does not assert independent reproduction.

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