# Controllable Summarization on MACSumDoc

Source: [Controllable Summarization on MACSumDoc](<https://www.sota2.com/research/sota/controllable-summarization-on-macsumdoc>)

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

Data updated: 2026-04-13T07:19:34.600Z

Evaluation of controllable summarization performance on the MACSumDoc dataset, utilizing metrics such as ROUGE, BERTScore, and control-specific metrics (Ext, Len, Spc, Top).

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 | Controllable Summarization |
| Dataset (including any stated split) | MACSumDoc |
| Metric | Ext Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 13 of 13 |
| 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

- [Ext Score](<https://www.sota2.com/research/sota/controllable-summarization-on-macsumdoc.md?metric=Ext%20Score>): Higher is better; unit: SCALAR.

- [Len Score](<https://www.sota2.com/research/sota/controllable-summarization-on-macsumdoc.md?metric=Len%20Score>): Higher is better; unit: SCALAR.

- [Spc Score](<https://www.sota2.com/research/sota/controllable-summarization-on-macsumdoc.md?metric=Spc%20Score>): Higher is better; unit: SCALAR.

- [Top-1 Score](<https://www.sota2.com/research/sota/controllable-summarization-on-macsumdoc.md?metric=Top-1%20Score>): Higher is better; unit: PERCENT.

- [ROUGE](<https://www.sota2.com/research/sota/controllable-summarization-on-macsumdoc.md?metric=ROUGE>): Higher is better; unit: PERCENT.

- [BERTScore](<https://www.sota2.com/research/sota/controllable-summarization-on-macsumdoc.md?metric=BERTScore>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Ext Score | Result date | Report paper |
| --- | --- | --- | --- |
| HP+SP (BARTlarge) | 9.04 | Not recorded | 68fa7bcb1e209131cea8a848 |
| Explicit self-planning | 8.22 | Not recorded | 68fa7bcb1e209131cea8a848 |
| Explicit self-planning+ | 8.19 | Not recorded | 68fa7bcb1e209131cea8a848 |
| Implicit self-planning | 8.14 | Not recorded | 68fa7bcb1e209131cea8a848 |
| Llama-3.3-Instruct | 8.03 | Not recorded | 68fa7bcb1e209131cea8a848 |
| Llama-3.2-Instruct | 7.84 | Not recorded | 68fa7bcb1e209131cea8a848 |
| PACO | 7.53 | Not recorded | 68fa7bcb1e209131cea8a848 |
| Qwen2.5-Instruct | 7.39 | Not recorded | 68fa7bcb1e209131cea8a848 |
| PACO | 6.37 | Not recorded | 68fa7bcb1e209131cea8a848 |
| Random sequential | 5.17 | Not recorded | 68fa7bcb1e209131cea8a848 |
| Joint-iterative | 4.58 | Not recorded | 68fa7bcb1e209131cea8a848 |
| PACO | 4.49 | Not recorded | 68fa7bcb1e209131cea8a848 |
| Reference summary | 0 | Not recorded | 68fa7bcb1e209131cea8a848 |

### HP+SP (BARTlarge)

Ext Score: 9.04. Result ID: 69dc98ab47f34ce924c32caa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fa51e209131cea46e0e.md>)

Recorded evaluation configuration: {"# of Params":"406M"}. This record does not assert independent reproduction.

### Explicit self-planning

Ext Score: 8.22. Result ID: 69dc98ab47f34ce924c32cb1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

Recorded evaluation configuration: {"# of Params":"70B","Base Model":"Llama-3.3-Instruct"}. This record does not assert independent reproduction.

### Explicit self-planning+

Ext Score: 8.19. Result ID: 69dc98ab47f34ce924c32cb2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

Recorded evaluation configuration: {"# of Params":"70B","Base Model":"Llama-3.3-Instruct"}. This record does not assert independent reproduction.

### Implicit self-planning

Ext Score: 8.14. Result ID: 69dc98ab47f34ce924c32cb0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

Recorded evaluation configuration: {"# of Params":"70B","Base Model":"Llama-3.3-Instruct"}. This record does not assert independent reproduction.

### Llama-3.3-Instruct

Ext Score: 8.03. Result ID: 69dc98ab47f34ce924c32caf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"# of Params":"70B"}. This record does not assert independent reproduction.

### Llama-3.2-Instruct

Ext Score: 7.84. Result ID: 69dc98ab47f34ce924c32cab.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"# of Params":"1B"}. This record does not assert independent reproduction.

### PACO

Ext Score: 7.53. Result ID: 69dc98ab47f34ce924c32cac.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

Recorded evaluation configuration: {"# of Params":"1B","Base Model":"Llama-3.2-Instruct"}. This record does not assert independent reproduction.

### Qwen2.5-Instruct

Ext Score: 7.39. Result ID: 69dc98ab47f34ce924c32cad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"# of Params":"7B"}. This record does not assert independent reproduction.

### PACO

Ext Score: 6.37. Result ID: 69dc98ab47f34ce924c32cae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

Recorded evaluation configuration: {"# of Params":"7B","Base Model":"Qwen2.5-Instruct"}. This record does not assert independent reproduction.

### Random sequential

Ext Score: 5.17. Result ID: 69dc98ab47f34ce924c32cb4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

Recorded evaluation configuration: {"# of Params":"70B","Base Model":"Llama-3.3-Instruct"}. This record does not assert independent reproduction.

### Joint-iterative

Ext Score: 4.58. Result ID: 69dc98ab47f34ce924c32cb3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

Recorded evaluation configuration: {"# of Params":"70B","Base Model":"Llama-3.3-Instruct"}. This record does not assert independent reproduction.

### PACO

Ext Score: 4.49. Result ID: 69dc98ab47f34ce924c32cb5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

Recorded evaluation configuration: {"# of Params":"70B","Base Model":"Llama-3.3-Instruct"}. This record does not assert independent reproduction.

### Reference summary

Ext Score: 0. Result ID: 69dc98ab47f34ce924c32ca9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcb1e209131cea8a848.md>)

Recorded evaluation configuration: {"# of Params":"-"}. This record does not assert independent reproduction.

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