# Long-term Memory Retrieval and Response Generation on LongMemEval

Source: [Long-term Memory Retrieval and Response Generation on LongMemEval](<https://www.sota2.com/research/sota/long-term-memory-retrieval-and-response-generation-on-longmemeval>)

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

Data updated: 2026-04-10T08:37:37.633Z

Evaluation of model performance on the LongMemEval dataset, specifically assessing long-term memory retrieval accuracy (Precision/Recall) and the relevance of the generated responses.

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 | Long-term Memory Retrieval and Response Generation |
| Dataset (including any stated split) | LongMemEval |
| Metric | Memory Precision |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [Memory Precision](<https://www.sota2.com/research/sota/long-term-memory-retrieval-and-response-generation-on-longmemeval.md?metric=Memory%20Precision>): Higher is better; unit: PERCENT.

- [Memory Recall](<https://www.sota2.com/research/sota/long-term-memory-retrieval-and-response-generation-on-longmemeval.md?metric=Memory%20Recall>): Higher is better; unit: PERCENT.

- [Response Relevance](<https://www.sota2.com/research/sota/long-term-memory-retrieval-and-response-generation-on-longmemeval.md?metric=Response%20Relevance>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Memory Precision | Result date | Report paper |
| --- | --- | --- | --- |
| EventWeave | 68.2 | Not recorded | 68fa77bf1e209131cea75138 |
| LD-Agent | 65.8 | Not recorded | 68fa77bf1e209131cea75138 |
| LifeLongMem | 63.4 | Not recorded | 68fa77bf1e209131cea75138 |

### EventWeave

Memory Precision: 68.2. Result ID: 69d8b6cc47f34ce924c19989.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77bf1e209131cea75138.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77bf1e209131cea75138.md>)

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

### LD-Agent

Memory Precision: 65.8. Result ID: 69d8b6cc47f34ce924c19988.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77bf1e209131cea75138.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76f41e209131cea703b0.md>)

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

### LifeLongMem

Memory Precision: 63.4. Result ID: 69d8b6cc47f34ce924c19987.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77bf1e209131cea75138.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c91e209131cea6f108.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.