# Cleft sentence structure analysis on CausalGym cleft Source: [Cleft sentence structure analysis on CausalGym cleft]() Collection index: [llms.txt]() Data updated: 2026-05-11T09:59:40.620Z Evaluation of model performance on analyzing cleft sentence structures using the CausalGym cleft subset, measured by Completeness, Selectivity, and Reliability. 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 | Cleft sentence structure analysis | | Dataset (including any stated split) | CausalGym cleft | | Metric | Completeness | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [Completeness](): Higher is better; unit: PERCENT. - [Selectivity](): Higher is better; unit: PERCENT. - [Reliability](): Higher is better; unit: PERCENT. ## Recorded results | Method | Completeness | Result date | Report paper | | --- | --- | --- | --- | | HDMI | 100 | Not recorded | 6a0162b5dbc836b87d8232c6 | | FGSM | 100 | Not recorded | 6a0162b5dbc836b87d8232c6 | | HDMI | 100 | Not recorded | 6a0162b5dbc836b87d8232c6 | | HDMI | 100 | Not recorded | 6a0162b5dbc836b87d8232c6 | | HDMI | 100 | Not recorded | 6a0162b5dbc836b87d8232c6 | | FGSM | 100 | Not recorded | 6a0162b5dbc836b87d8232c6 | | PGD | 99.46 | Not recorded | 6a0162b5dbc836b87d8232c6 | | PGD | 99.46 | Not recorded | 6a0162b5dbc836b87d8232c6 | | PGD | 95.99 | Not recorded | 6a0162b5dbc836b87d8232c6 | | PGD | 95.99 | Not recorded | 6a0162b5dbc836b87d8232c6 | | AlterRep | 86.03 | Not recorded | 6a0162b5dbc836b87d8232c6 | | AlterRep | 86.03 | Not recorded | 6a0162b5dbc836b87d8232c6 | | FGSM | 79 | Not recorded | 6a0162b5dbc836b87d8232c6 | | FGSM | 79 | Not recorded | 6a0162b5dbc836b87d8232c6 | | AlterRep | 73 | Not recorded | 6a0162b5dbc836b87d8232c6 | | AlterRep | 73 | Not recorded | 6a0162b5dbc836b87d8232c6 | ### HDMI Completeness: 100. Result ID: 6a01a86547f34ce924e40649. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Meta-Llama-3-8B-Instruct"}. This record does not assert independent reproduction. ### FGSM Completeness: 100. Result ID: 6a01a86547f34ce924e4064b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Meta-Llama-3-8B-Instruct"}. This record does not assert independent reproduction. ### HDMI Completeness: 100. Result ID: 6a01a86547f34ce924e4064d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EleutherAI/pythia-70m"}. This record does not assert independent reproduction. ### HDMI Completeness: 100. Result ID: 6a01a86547f34ce924e40691. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Meta-Llama-3-8B-Instruct"}. This record does not assert independent reproduction. ### HDMI Completeness: 100. Result ID: 6a01a86547f34ce924e40692. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EleutherAI/pythia-70m"}. This record does not assert independent reproduction. ### FGSM Completeness: 100. Result ID: 6a01a86547f34ce924e40695. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Meta-Llama-3-8B-Instruct"}. This record does not assert independent reproduction. ### PGD Completeness: 99.46. Result ID: 6a01a86547f34ce924e4064c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Meta-Llama-3-8B-Instruct"}. This record does not assert independent reproduction. ### PGD Completeness: 99.46. Result ID: 6a01a86547f34ce924e40697. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Meta-Llama-3-8B-Instruct"}. This record does not assert independent reproduction. ### PGD Completeness: 95.99. Result ID: 6a01a86547f34ce924e40650. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EleutherAI/pythia-70m"}. This record does not assert independent reproduction. ### PGD Completeness: 95.99. Result ID: 6a01a86547f34ce924e40698. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EleutherAI/pythia-70m"}. This record does not assert independent reproduction. ### AlterRep Completeness: 86.03. Result ID: 6a01a86547f34ce924e4064a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Meta-Llama-3-8B-Instruct"}. This record does not assert independent reproduction. ### AlterRep Completeness: 86.03. Result ID: 6a01a86547f34ce924e40693. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Meta-Llama-3-8B-Instruct"}. This record does not assert independent reproduction. ### FGSM Completeness: 79. Result ID: 6a01a86547f34ce924e4064f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EleutherAI/pythia-70m"}. This record does not assert independent reproduction. ### FGSM Completeness: 79. Result ID: 6a01a86547f34ce924e40696. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EleutherAI/pythia-70m"}. This record does not assert independent reproduction. ### AlterRep Completeness: 73. Result ID: 6a01a86547f34ce924e4064e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EleutherAI/pythia-70m"}. This record does not assert independent reproduction. ### AlterRep Completeness: 73. Result ID: 6a01a86547f34ce924e40694. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EleutherAI/pythia-70m"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.