# Long Multi-session Question Answering on LongMemEval 1.0 (test)

Source: [Long Multi-session Question Answering on LongMemEval 1.0 (test)](<https://www.sota2.com/research/sota/long-multi-session-question-answering-on-longmemeval-1-0-test>)

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

Data updated: 2026-06-11T10:26:30.425Z

Evaluation of Long Multi-session Question Answering performance on the test split of LongMemEval version 1.0, using F1-based metrics including KU F1, SS-assist F1, and Overall F1.

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 Multi-session Question Answering |
| Dataset (including any stated split) | LongMemEval 1.0 (test) |
| Metric | KU F1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [KU F1](<https://www.sota2.com/research/sota/long-multi-session-question-answering-on-longmemeval-1-0-test.md?metric=KU%20F1>): Higher is better; unit: PERCENT.

- [SS-assist F1](<https://www.sota2.com/research/sota/long-multi-session-question-answering-on-longmemeval-1-0-test.md?metric=SS-assist%20F1>): Higher is better; unit: PERCENT.

- [SS-prefer F1](<https://www.sota2.com/research/sota/long-multi-session-question-answering-on-longmemeval-1-0-test.md?metric=SS-prefer%20F1>): Higher is better; unit: PERCENT.

- [SS-user F1](<https://www.sota2.com/research/sota/long-multi-session-question-answering-on-longmemeval-1-0-test.md?metric=SS-user%20F1>): Higher is better; unit: PERCENT.

- [Temporal F1](<https://www.sota2.com/research/sota/long-multi-session-question-answering-on-longmemeval-1-0-test.md?metric=Temporal%20F1>): Higher is better; unit: PERCENT.

- [Overall F1](<https://www.sota2.com/research/sota/long-multi-session-question-answering-on-longmemeval-1-0-test.md?metric=Overall%20F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | KU F1 | Result date | Report paper |
| --- | --- | --- | --- |
| HORMA | 34.6 | Not recorded | 6a2a4136d33dfb23e783943b |
| Acon | 24.5 | Not recorded | 6a2a4136d33dfb23e783943b |
| ReSum | 23.2 | Not recorded | 6a2a4136d33dfb23e783943b |
| A-MEM | 21.7 | Not recorded | 6a2a4136d33dfb23e783943b |
| Truncation | 21.4 | Not recorded | 6a2a4136d33dfb23e783943b |
| Mem0 | 16.3 | Not recorded | 6a2a4136d33dfb23e783943b |
| Embedding Retrieval | 13.7 | Not recorded | 6a2a4136d33dfb23e783943b |
| No limit | 9.2 | Not recorded | 6a2a4136d33dfb23e783943b |
| Embedding | 8.8 | Not recorded | 6a2a4136d33dfb23e783943b |
| Slide Window | 7.1 | Not recorded | 6a2a4136d33dfb23e783943b |

### HORMA

KU F1: 34.6. Result ID: 6a2a8bedfbc373592c285770.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"50K tokens","Memory Architecture":"External Dynamic Memory"}. This record does not assert independent reproduction.

### Acon

KU F1: 24.5. Result ID: 6a2a8bedfbc373592c28576c.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c441e209131cea8d495.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"50K tokens","Memory Architecture":"Internal Dynamic Memory"}. This record does not assert independent reproduction.

### ReSum

KU F1: 23.2. Result ID: 6a2a8bedfbc373592c28576b.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c7b5.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"50K tokens","Memory Architecture":"Internal Dynamic Memory"}. This record does not assert independent reproduction.

### A-MEM

KU F1: 21.7. Result ID: 6a2a8bedfbc373592c28576d.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bf31e209131cea8b869.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"50K tokens","Memory Architecture":"External Dynamic Memory"}. This record does not assert independent reproduction.

### Truncation

KU F1: 21.4. Result ID: 6a2a8bedfbc373592c285768.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"50K tokens","Memory Architecture":"Internal Static Memory"}. This record does not assert independent reproduction.

### Mem0

KU F1: 16.3. Result ID: 6a2a8bedfbc373592c28576e.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa784b1e209131cea77ad6.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"50K tokens","Memory Architecture":"External Dynamic Memory"}. This record does not assert independent reproduction.

### Embedding Retrieval

KU F1: 13.7. Result ID: 6a2a8bedfbc373592c28576f.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"50K tokens","Memory Architecture":"External Dynamic Memory"}. This record does not assert independent reproduction.

### No limit

KU F1: 9.2. Result ID: 6a2a8bedfbc373592c285767.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"Unconstrained"}. This record does not assert independent reproduction.

### Embedding

KU F1: 8.8. Result ID: 6a2a8bedfbc373592c28576a.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"50K tokens","Memory Architecture":"Internal Static Memory"}. This record does not assert independent reproduction.

### Slide Window

KU F1: 7.1. Result ID: 6a2a8bedfbc373592c285769.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a4136d33dfb23e783943b.md>)

Recorded evaluation configuration: {"Backbone":"Claude Sonnet 4.5","Context Limit":"50K tokens","Memory Architecture":"Internal Static Memory"}. This record does not assert independent reproduction.

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