# MAS Test-Time Scaling on HotpotQA

Source: [MAS Test-Time Scaling on HotpotQA](<https://www.sota2.com/research/sota/mas-test-time-scaling-on-hotpotqa>)

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

Data updated: 2026-06-12T08:22:02.820Z

Evaluation of model performance on the HotpotQA dataset specifically measuring accuracy and computational scaling (tokens used during testing) under MAS Test-Time Scaling conditions.

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 | MAS Test-Time Scaling |
| Dataset (including any stated split) | HotpotQA |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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/mas-test-time-scaling-on-hotpotqa.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Tokens Used (M)](<https://www.sota2.com/research/sota/mas-test-time-scaling-on-hotpotqa.md?metric=Tokens%20Used%20(M)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Oracle (UB) | 57 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| GPT-5.4-mini | 46 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| GPT-5-mini | 44 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| MAS-Orchestra (LB) | 42.5 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| Maj. Vote | 42.5 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| Orch-RM | 42.5 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| GPT-4.1-mini | 42 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| GPT-5.2 | 42 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| Skywork-V2-Qwen3-4B | 42 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| Skywork-V2-Llama-3.2-3B | 41.5 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| Skywork-V2-Qwen3-8B | 41 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| GPT-5-mini | 40.5 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| Skywork-LLaMA-3.1-8B | 40.5 | Not recorded | 6a2b92d4d33dfb23e783b2dd |
| log P | 39.5 | Not recorded | 6a2b92d4d33dfb23e783b2dd |

### Oracle (UB)

Accuracy: 57. Result ID: 6a2bc196fbc373592c2a302a.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

Recorded evaluation configuration: {"Verification Level":"Trajectory","N":"8"}. This record does not assert independent reproduction.

### GPT-5.4-mini

Accuracy: 46. Result ID: 6a2bc196fbc373592c2a302e.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695f39f053fdd5c361743ecb.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

### GPT-5-mini

Accuracy: 44. Result ID: 6a2bc196fbc373592c2a3029.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695f39f053fdd5c361743ecb.md>)

Recorded evaluation configuration: {"Verification Level":"Trajectory","N":"8"}. This record does not assert independent reproduction.

### MAS-Orchestra (LB)

Accuracy: 42.5. Result ID: 6a2bc196fbc373592c2a3027.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6971af07e25ad2fa9d1ed973.md>)

Recorded evaluation configuration: {"Verification Level":"-","N":"8"}. This record does not assert independent reproduction.

### Maj. Vote

Accuracy: 42.5. Result ID: 6a2bc196fbc373592c2a3028.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb61e209131cea41b33.md>)

Recorded evaluation configuration: {"Verification Level":"Trajectory","N":"8"}. This record does not assert independent reproduction.

### Orch-RM

Accuracy: 42.5. Result ID: 6a2bc196fbc373592c2a3034.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

### GPT-4.1-mini

Accuracy: 42. Result ID: 6a2bc196fbc373592c2a302b.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

### GPT-5.2

Accuracy: 42. Result ID: 6a2bc196fbc373592c2a302d.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695f39f053fdd5c361743ecb.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

### Skywork-V2-Qwen3-4B

Accuracy: 42. Result ID: 6a2bc196fbc373592c2a3032.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79c61e209131cea7f4fe.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

### Skywork-V2-Llama-3.2-3B

Accuracy: 41.5. Result ID: 6a2bc196fbc373592c2a3031.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79c61e209131cea7f4fe.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

### Skywork-V2-Qwen3-8B

Accuracy: 41. Result ID: 6a2bc196fbc373592c2a3033.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79c61e209131cea7f4fe.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

### GPT-5-mini

Accuracy: 40.5. Result ID: 6a2bc196fbc373592c2a302c.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695f39f053fdd5c361743ecb.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

### Skywork-LLaMA-3.1-8B

Accuracy: 40.5. Result ID: 6a2bc196fbc373592c2a3030.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa755d1e209131cea67806.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

### log P

Accuracy: 39.5. Result ID: 6a2bc196fbc373592c2a302f.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d4d33dfb23e783b2dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5dfa3.md>)

Recorded evaluation configuration: {"Verification Level":"Orchestration","N":"8"}. This record does not assert independent reproduction.

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