# Implicit-conflict resolution on STALE 1.0 (full)

Source: [Implicit-conflict resolution on STALE 1.0 (full)](<https://www.sota2.com/research/sota/implicit-conflict-resolution-on-stale-1-0-full>)

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

Data updated: 2026-05-08T11:20:41.041Z

Evaluation of model performance on implicit conflict resolution tasks using the full set of the STALE 1.0 dataset, measured by Type I/II SR, PR, and IPA Accuracies, and Overall Accuracy.

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 | Implicit-conflict resolution |
| Dataset (including any stated split) | STALE 1.0 (full) |
| Metric | Type I SR Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Type I SR Accuracy](<https://www.sota2.com/research/sota/implicit-conflict-resolution-on-stale-1-0-full.md?metric=Type%20I%20SR%20Accuracy>): Higher is better; unit: PERCENT.

- [Type I PR Accuracy](<https://www.sota2.com/research/sota/implicit-conflict-resolution-on-stale-1-0-full.md?metric=Type%20I%20PR%20Accuracy>): Higher is better; unit: PERCENT.

- [Type I IPA Accuracy](<https://www.sota2.com/research/sota/implicit-conflict-resolution-on-stale-1-0-full.md?metric=Type%20I%20IPA%20Accuracy>): Higher is better; unit: PERCENT.

- [Type II SR Accuracy](<https://www.sota2.com/research/sota/implicit-conflict-resolution-on-stale-1-0-full.md?metric=Type%20II%20SR%20Accuracy>): Higher is better; unit: PERCENT.

- [Type II PR Accuracy](<https://www.sota2.com/research/sota/implicit-conflict-resolution-on-stale-1-0-full.md?metric=Type%20II%20PR%20Accuracy>): Higher is better; unit: PERCENT.

- [Type II IPA Accuracy](<https://www.sota2.com/research/sota/implicit-conflict-resolution-on-stale-1-0-full.md?metric=Type%20II%20IPA%20Accuracy>): Higher is better; unit: PERCENT.

- [Overall Accuracy](<https://www.sota2.com/research/sota/implicit-conflict-resolution-on-stale-1-0-full.md?metric=Overall%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Type I SR Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Gemini-3.1-pro | 92 | Not recorded | 69fd6e5e5497b3b670c8975d |
| CUPMEM | 91 | Not recorded | 69fd6e5e5497b3b670c8975d |
| Qwen3.5-27B | 76 | Not recorded | 69fd6e5e5497b3b670c8975d |
| LightMem | 52.5 | Not recorded | 69fd6e5e5497b3b670c8975d |
| Gemini-3.1-flash-lite | 41 | Not recorded | 69fd6e5e5497b3b670c8975d |
| Qwen3.5-9B | 36 | Not recorded | 69fd6e5e5497b3b670c8975d |
| GPT-5.4 | 35 | Not recorded | 69fd6e5e5497b3b670c8975d |
| GPT-4o-mini\* | 30 | Not recorded | 69fd6e5e5497b3b670c8975d |
| GPT-5.4-nano | 20.5 | Not recorded | 69fd6e5e5497b3b670c8975d |
| mem-0 | 17 | Not recorded | 69fd6e5e5497b3b670c8975d |
| LiCoMemory | 15.5 | Not recorded | 69fd6e5e5497b3b670c8975d |
| A-mem | 13.5 | Not recorded | 69fd6e5e5497b3b670c8975d |
| MiniMax-M2.5 | 10.5 | Not recorded | 69fd6e5e5497b3b670c8975d |
| Zep | 10 | Not recorded | 69fd6e5e5497b3b670c8975d |
| Llama-3.3-70B-Instruct\* | 6.5 | Not recorded | 69fd6e5e5497b3b670c8975d |

### Gemini-3.1-pro

Type I SR Accuracy: 92. Result ID: 69fdc6fc47f34ce924e1b0a0.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Closed-source LLMs"}. This record does not assert independent reproduction.

### CUPMEM

Type I SR Accuracy: 91. Result ID: 69fdc6fc47f34ce924e1b0aa.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Memory Frameworks"}. This record does not assert independent reproduction.

### Qwen3.5-27B

Type I SR Accuracy: 76. Result ID: 69fdc6fc47f34ce924e1b0a3.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Open-source LLMs"}. This record does not assert independent reproduction.

### LightMem

Type I SR Accuracy: 52.5. Result ID: 69fdc6fc47f34ce924e1b0a5.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c6d1e209131cea8e04d.md>)

Recorded evaluation configuration: {"Category":"Memory Frameworks"}. This record does not assert independent reproduction.

### Gemini-3.1-flash-lite

Type I SR Accuracy: 41. Result ID: 69fdc6fc47f34ce924e1b09f.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Closed-source LLMs"}. This record does not assert independent reproduction.

### Qwen3.5-9B

Type I SR Accuracy: 36. Result ID: 69fdc6fc47f34ce924e1b0a2.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Open-source LLMs"}. This record does not assert independent reproduction.

### GPT-5.4

Type I SR Accuracy: 35. Result ID: 69fdc6fc47f34ce924e1b09e.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Closed-source LLMs"}. This record does not assert independent reproduction.

### GPT-4o-mini\*

Type I SR Accuracy: 30. Result ID: 69fdc6fc47f34ce924e1b09c.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Closed-source LLMs"}. This record does not assert independent reproduction.

### GPT-5.4-nano

Type I SR Accuracy: 20.5. Result ID: 69fdc6fc47f34ce924e1b09d.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Closed-source LLMs"}. This record does not assert independent reproduction.

### mem-0

Type I SR Accuracy: 17. Result ID: 69fdc6fc47f34ce924e1b0a9.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

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

Recorded evaluation configuration: {"Category":"Memory Frameworks"}. This record does not assert independent reproduction.

### LiCoMemory

Type I SR Accuracy: 15.5. Result ID: 69fdc6fc47f34ce924e1b0a7.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/690988863ef7ccd74e5e5c5d.md>)

Recorded evaluation configuration: {"Category":"Memory Frameworks"}. This record does not assert independent reproduction.

### A-mem

Type I SR Accuracy: 13.5. Result ID: 69fdc6fc47f34ce924e1b0a8.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

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

Recorded evaluation configuration: {"Category":"Memory Frameworks"}. This record does not assert independent reproduction.

### MiniMax-M2.5

Type I SR Accuracy: 10.5. Result ID: 69fdc6fc47f34ce924e1b0a4.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Open-source LLMs"}. This record does not assert independent reproduction.

### Zep

Type I SR Accuracy: 10. Result ID: 69fdc6fc47f34ce924e1b0a6.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76b41e209131cea6e944.md>)

Recorded evaluation configuration: {"Category":"Memory Frameworks"}. This record does not assert independent reproduction.

### Llama-3.3-70B-Instruct\*

Type I SR Accuracy: 6.5. Result ID: 69fdc6fc47f34ce924e1b0a1.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5e5497b3b670c8975d.md>)

Recorded evaluation configuration: {"Category":"Open-source LLMs"}. This record does not assert independent reproduction.

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