# POMDP Simulation on Tag

Source: [POMDP Simulation on Tag](<https://www.sota2.com/research/sota/pomdp-simulation-on-tag>)

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

Data updated: 2026-02-26T06:02:15.563Z

Evaluation of models simulating Partially Observable Markov Decision Processes (POMDP) environments using the Tag dataset, focusing on reward metrics and suggestion counts.

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 | POMDP Simulation |
| Dataset (including any stated split) | Tag |
| Metric | Reward |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| 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

- [Reward](<https://www.sota2.com/research/sota/pomdp-simulation-on-tag.md?metric=Reward>): Higher is better; unit: SCALAR.

- [Suggestion Count](<https://www.sota2.com/research/sota/pomdp-simulation-on-tag.md?metric=Suggestion%20Count>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Reward | Result date | Report paper |
| --- | --- | --- | --- |
| Perfect | 1.7 | Not recorded | 68fa6d801e209131cea3a1bc |
| Naive | -1.6 | Not recorded | 68fa6d801e209131cea3a1bc |
| Scaled Agent | -1.8 | Not recorded | 68fa6d801e209131cea3a1bc |
| Noisy Agent | -1.8 | Not recorded | 68fa6d801e209131cea3a1bc |
| Noisy Agent | -2 | Not recorded | 68fa6d801e209131cea3a1bc |
| Scaled Agent | -2.4 | Not recorded | 68fa6d801e209131cea3a1bc |
| Noisy Agent | -2.4 | Not recorded | 68fa6d801e209131cea3a1bc |
| Scaled Agent | -3.6 | Not recorded | 68fa6d801e209131cea3a1bc |
| Naive | -3.8 | Not recorded | 68fa6d801e209131cea3a1bc |
| Naive | -6.8 | Not recorded | 68fa6d801e209131cea3a1bc |
| Normal | -10.7 | Not recorded | 68fa6d801e209131cea3a1bc |

### Perfect

Reward: 1.7. Result ID: 698fb5a33a6c314849fb4cec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

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

### Naive

Reward: -1.6. Result ID: 698fb5a33a6c314849fb4ced.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

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

### Scaled Agent

Reward: -1.8. Result ID: 698fb5a33a6c314849fb4cf0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Implementation](<https://github.com/sisl/action_suggestions>)

Recorded evaluation configuration: {"tau":0.99}. This record does not assert independent reproduction.

### Noisy Agent

Reward: -1.8. Result ID: 698fb5a33a6c314849fb4cf3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Implementation](<https://github.com/sisl/action_suggestions>)

Recorded evaluation configuration: {"lambda":5}. This record does not assert independent reproduction.

### Noisy Agent

Reward: -2. Result ID: 698fb5a33a6c314849fb4cf4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Implementation](<https://github.com/sisl/action_suggestions>)

Recorded evaluation configuration: {"lambda":2}. This record does not assert independent reproduction.

### Scaled Agent

Reward: -2.4. Result ID: 698fb5a33a6c314849fb4cf1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Implementation](<https://github.com/sisl/action_suggestions>)

Recorded evaluation configuration: {"tau":0.75}. This record does not assert independent reproduction.

### Noisy Agent

Reward: -2.4. Result ID: 698fb5a33a6c314849fb4cf5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Implementation](<https://github.com/sisl/action_suggestions>)

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

### Scaled Agent

Reward: -3.6. Result ID: 698fb5a33a6c314849fb4cf2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

[Implementation](<https://github.com/sisl/action_suggestions>)

Recorded evaluation configuration: {"tau":0.5}. This record does not assert independent reproduction.

### Naive

Reward: -3.8. Result ID: 698fb5a33a6c314849fb4cee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

Recorded evaluation configuration: {"v":0.75}. This record does not assert independent reproduction.

### Naive

Reward: -6.8. Result ID: 698fb5a33a6c314849fb4cef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

Recorded evaluation configuration: {"v":0.5}. This record does not assert independent reproduction.

### Normal

Reward: -10.7. Result ID: 698fb5a33a6c314849fb4ceb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d801e209131cea3a1bc.md>)

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

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