# Multi-hop Grounding on CorpusQA 1M

Source: [Multi-hop Grounding on CorpusQA 1M](<https://www.sota2.com/research/sota/multi-hop-grounding-on-corpusqa-1m>)

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

Data updated: 2026-02-26T06:03:54.095Z

Evaluation of model performance on the CorpusQA 1M dataset for multi-hop grounding tasks.

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 | Multi-hop grounding |
| Dataset (including any stated split) | CorpusQA 1M |
| Metric | Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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/multi-hop-grounding-on-corpusqa-1m.md?metric=Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Score | Result date | Report paper |
| --- | --- | --- | --- |
| Gemini-2.5-Pro | 53.11 | Not recorded | 6940e7d8456146936284fc0a |
| Gemini-2.5-Flash-Thinking | 36.91 | Not recorded | 6940e7d8456146936284fc0a |
| QwenLong-L1.5-30B-A3B | 20.72 | Not recorded | 6940e7d8456146936284fc0a |
| Qwen3-30B-A3B-Thinking-2507 | 15.32 | Not recorded | 6940e7d8456146936284fc0a |
| MemAgent-14B | 9.7 | Not recorded | 6940e7d8456146936284fc0a |
| Qwen-Flash-Thinking-1M | 2.75 | Not recorded | 6940e7d8456146936284fc0a |

### Gemini-2.5-Pro

Score: 53.11. Result ID: 6996c0b65698665ec40bdeb1.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7d8456146936284fc0a.md>)

Recorded evaluation configuration: {"Evaluation Framework":"Full-context Inference"}. This record does not assert independent reproduction.

### Gemini-2.5-Flash-Thinking

Score: 36.91. Result ID: 6996c0b65698665ec40bdeb2.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7d8456146936284fc0a.md>)

Recorded evaluation configuration: {"Evaluation Framework":"Full-context Inference"}. This record does not assert independent reproduction.

### QwenLong-L1.5-30B-A3B

Score: 20.72. Result ID: 6996c0b65698665ec40bdeb6.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7d8456146936284fc0a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6940e7d8456146936284fc0a.md>)

[Implementation](<https://github.com/Tongyi-Zhiwen/Qwen-Doc>)

Recorded evaluation configuration: {"Evaluation Framework":"Memory Agent Framework"}. This record does not assert independent reproduction.

### Qwen3-30B-A3B-Thinking-2507

Score: 15.32. Result ID: 6996c0b65698665ec40bdeb5.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7d8456146936284fc0a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Evaluation Framework":"Memory Agent Framework"}. This record does not assert independent reproduction.

### MemAgent-14B

Score: 9.7. Result ID: 6996c0b65698665ec40bdeb4.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7d8456146936284fc0a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79c71e209131cea7f569.md>)

Recorded evaluation configuration: {"Evaluation Framework":"Memory Agent Framework"}. This record does not assert independent reproduction.

### Qwen-Flash-Thinking-1M

Score: 2.75. Result ID: 6996c0b65698665ec40bdeb3.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7d8456146936284fc0a.md>)

Recorded evaluation configuration: {"Evaluation Framework":"Full-context Inference"}. This record does not assert independent reproduction.

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