# Sequential-insert knowledge insertion on CrossCult-KIBench 1.0 (test)

Source: [Sequential-insert knowledge insertion on CrossCult-KIBench 1.0 (test)](<https://www.sota2.com/research/sota/sequential-insert-knowledge-insertion-on-crosscult-kibench-1-0-test>)

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

Data updated: 2026-05-08T10:07:57.180Z

Evaluation of sequential knowledge insertion capabilities on the test split of the CrossCult-KIBench 1.0 dataset, measured using ROUGE-L and LLM Judge scores across relational, generative, and locational aspects.

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 | Sequential-insert knowledge insertion |
| Dataset (including any stated split) | CrossCult-KIBench 1.0 (test) |
| Metric | Final Relational ROUGE-L |
| Unit | SCALAR |
| 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

- [Final Relational ROUGE-L](<https://www.sota2.com/research/sota/sequential-insert-knowledge-insertion-on-crosscult-kibench-1-0-test.md?metric=Final%20Relational%20ROUGE-L>): Higher is better; unit: SCALAR.

- [Final Relational LLM Judge Score](<https://www.sota2.com/research/sota/sequential-insert-knowledge-insertion-on-crosscult-kibench-1-0-test.md?metric=Final%20Relational%20LLM%20Judge%20Score>): Higher is better; unit: SCALAR.

- [Final Generative ROUGE-L](<https://www.sota2.com/research/sota/sequential-insert-knowledge-insertion-on-crosscult-kibench-1-0-test.md?metric=Final%20Generative%20ROUGE-L>): Higher is better; unit: SCALAR.

- [Final Generative LLM Judge Score](<https://www.sota2.com/research/sota/sequential-insert-knowledge-insertion-on-crosscult-kibench-1-0-test.md?metric=Final%20Generative%20LLM%20Judge%20Score>): Higher is better; unit: SCALAR.

- [Final Locational ROUGE-L](<https://www.sota2.com/research/sota/sequential-insert-knowledge-insertion-on-crosscult-kibench-1-0-test.md?metric=Final%20Locational%20ROUGE-L>): Higher is better; unit: SCALAR.

- [Final Locational LLM Judge Score](<https://www.sota2.com/research/sota/sequential-insert-knowledge-insertion-on-crosscult-kibench-1-0-test.md?metric=Final%20Locational%20LLM%20Judge%20Score>): Higher is better; unit: SCALAR.

- [Overall ROUGE-L](<https://www.sota2.com/research/sota/sequential-insert-knowledge-insertion-on-crosscult-kibench-1-0-test.md?metric=Overall%20ROUGE-L>): Higher is better; unit: SCALAR.

- [Overall LLM Judge Score](<https://www.sota2.com/research/sota/sequential-insert-knowledge-insertion-on-crosscult-kibench-1-0-test.md?metric=Overall%20LLM%20Judge%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Final Relational ROUGE-L | Result date | Report paper |
| --- | --- | --- | --- |
| MCKI | 85.57 | Not recorded | 69fd6e165497b3b670c89424 |
| IKE | 75.44 | Not recorded | 69fd6e165497b3b670c89424 |
| MSCKE | 66.08 | Not recorded | 69fd6e165497b3b670c89424 |
| MSCKE | 62.1 | Not recorded | 69fd6e165497b3b670c89424 |
| MCKI | 59.83 | Not recorded | 69fd6e165497b3b670c89424 |
| SERAC | 46.62 | Not recorded | 69fd6e165497b3b670c89424 |
| SERAC | 36.23 | Not recorded | 69fd6e165497b3b670c89424 |
| IKE | 13.85 | Not recorded | 69fd6e165497b3b670c89424 |
| Base Model | 13.55 | Not recorded | 69fd6e165497b3b670c89424 |
| Base Model | 12.1 | Not recorded | 69fd6e165497b3b670c89424 |
| FineTune | 6.8 | Not recorded | 69fd6e165497b3b670c89424 |
| MEND | 4.07 | Not recorded | 69fd6e165497b3b670c89424 |
| MEND | 2.42 | Not recorded | 69fd6e165497b3b670c89424 |
| FineTune | 2.37 | Not recorded | 69fd6e165497b3b670c89424 |

### MCKI

Final Relational ROUGE-L: 85.57. Result ID: 69fdb52247f34ce924e12d55.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

Recorded evaluation configuration: {"Base Model":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### IKE

Final Relational ROUGE-L: 75.44. Result ID: 69fdb52247f34ce924e12d51.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f651e209131cea45479.md>)

Recorded evaluation configuration: {"Base Model":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### MSCKE

Final Relational ROUGE-L: 66.08. Result ID: 69fdb52247f34ce924e12d54.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75c01e209131cea69b54.md>)

Recorded evaluation configuration: {"Base Model":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### MSCKE

Final Relational ROUGE-L: 62.1. Result ID: 69fdb52247f34ce924e12d5b.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75c01e209131cea69b54.md>)

Recorded evaluation configuration: {"Base Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### MCKI

Final Relational ROUGE-L: 59.83. Result ID: 69fdb52247f34ce924e12d5c.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

Recorded evaluation configuration: {"Base Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### SERAC

Final Relational ROUGE-L: 46.62. Result ID: 69fdb52247f34ce924e12d53.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cc71e209131cea35f07.md>)

Recorded evaluation configuration: {"Base Model":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### SERAC

Final Relational ROUGE-L: 36.23. Result ID: 69fdb52247f34ce924e12d5a.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cc71e209131cea35f07.md>)

Recorded evaluation configuration: {"Base Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### IKE

Final Relational ROUGE-L: 13.85. Result ID: 69fdb52247f34ce924e12d58.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f651e209131cea45479.md>)

Recorded evaluation configuration: {"Base Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### Base Model

Final Relational ROUGE-L: 13.55. Result ID: 69fdb52247f34ce924e12d4f.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

Recorded evaluation configuration: {"Base Model":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### Base Model

Final Relational ROUGE-L: 12.1. Result ID: 69fdb52247f34ce924e12d56.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

Recorded evaluation configuration: {"Base Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### FineTune

Final Relational ROUGE-L: 6.8. Result ID: 69fdb52247f34ce924e12d50.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

Recorded evaluation configuration: {"Base Model":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### MEND

Final Relational ROUGE-L: 4.07. Result ID: 69fdb52247f34ce924e12d59.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cc61e209131cea35ea9.md>)

Recorded evaluation configuration: {"Base Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### MEND

Final Relational ROUGE-L: 2.42. Result ID: 69fdb52247f34ce924e12d52.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cc61e209131cea35ea9.md>)

Recorded evaluation configuration: {"Base Model":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### FineTune

Final Relational ROUGE-L: 2.37. Result ID: 69fdb52247f34ce924e12d57.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e165497b3b670c89424.md>)

Recorded evaluation configuration: {"Base Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

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