# Commonsense Question Answering on ECQA (test) Source: [Commonsense Question Answering on ECQA (test)]() Collection index: [llms.txt]() Data updated: 2026-05-18T07:54:53.065Z Evaluation of model performance on the test split of the ECQA dataset for Commonsense Question Answering, focusing on accuracy and faithfulness metrics. 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 | Commonsense Question Answering | | Dataset (including any stated split) | ECQA (test) | | Metric | Accuracy | | Unit | PERCENT | | 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 - [Accuracy](): Higher is better; unit: PERCENT. - [CT Unfaithfulness](): Lower is better; unit: PERCENT. - [CCT Faithfulness](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Llama2-70B | 79.7 | Not recorded | 68fa738f1e209131cea5d3bb | | Llama2-70B | 77.8 | Not recorded | 68fa738f1e209131cea5d3bb | | NV-Embed | 71.75 | Not recorded | 6a0a9cdddbc836b87d82b22f | | Llama2-13B | 71.4 | Not recorded | 68fa738f1e209131cea5d3bb | | LESS | 71.19 | Not recorded | 6a0a9cdddbc836b87d82b22f | | Interaction-Aware Influence Function (Ours) | 71.17 | Not recorded | 6a0a9cdddbc836b87d82b22f | | RDS+ | 70.88 | Not recorded | 6a0a9cdddbc836b87d82b22f | | Random | 69.85 | Not recorded | 6a0a9cdddbc836b87d82b22f | | Additive IF | 69.52 | Not recorded | 6a0a9cdddbc836b87d82b22f | | Llama2-13B | 68 | Not recorded | 68fa738f1e209131cea5d3bb | | Llama2-7B | 55.2 | Not recorded | 68fa738f1e209131cea5d3bb | | Llama2-7B | 54.1 | Not recorded | 68fa738f1e209131cea5d3bb | | Random | 20 | Not recorded | 68fa738f1e209131cea5d3bb | ### Llama2-70B Accuracy: 79.7. Result ID: 6992f1e53a6c314849128d19. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Prompt Order":"predict-then-explain (PE)"}. This record does not assert independent reproduction. ### Llama2-70B Accuracy: 77.8. Result ID: 6992f1e53a6c314849128d1a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Prompt Order":"explain-then-predict (EP)"}. This record does not assert independent reproduction. ### NV-Embed Accuracy: 71.75. Result ID: 6a0ac5ca47f34ce924f0ab57. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1-8B","Evaluation Protocol":"Fine-tuned","Number of Seeds":"5"}. This record does not assert independent reproduction. ### Llama2-13B Accuracy: 71.4. Result ID: 6992f1e53a6c314849128d18. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Prompt Order":"explain-then-predict (EP)"}. This record does not assert independent reproduction. ### LESS Accuracy: 71.19. Result ID: 6a0ac5ca47f34ce924f0ab55. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1-8B","Evaluation Protocol":"Fine-tuned","Number of Seeds":"5"}. This record does not assert independent reproduction. ### Interaction-Aware Influence Function (Ours) Accuracy: 71.17. Result ID: 6a0ac5ca47f34ce924f0ab58. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1-8B","Evaluation Protocol":"Fine-tuned","Number of Seeds":"5"}. This record does not assert independent reproduction. ### RDS+ Accuracy: 70.88. Result ID: 6a0ac5ca47f34ce924f0ab56. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1-8B","Evaluation Protocol":"Fine-tuned","Number of Seeds":"5"}. This record does not assert independent reproduction. ### Random Accuracy: 69.85. Result ID: 6a0ac5ca47f34ce924f0ab53. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1-8B","Evaluation Protocol":"Fine-tuned","Number of Seeds":"5"}. This record does not assert independent reproduction. ### Additive IF Accuracy: 69.52. Result ID: 6a0ac5ca47f34ce924f0ab54. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1-8B","Evaluation Protocol":"Fine-tuned","Number of Seeds":"5"}. This record does not assert independent reproduction. ### Llama2-13B Accuracy: 68. Result ID: 6992f1e53a6c314849128d17. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Prompt Order":"predict-then-explain (PE)"}. This record does not assert independent reproduction. ### Llama2-7B Accuracy: 55.2. Result ID: 6992f1e53a6c314849128d16. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Prompt Order":"explain-then-predict (EP)"}. This record does not assert independent reproduction. ### Llama2-7B Accuracy: 54.1. Result ID: 6992f1e53a6c314849128d15. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Prompt Order":"predict-then-explain (PE)"}. This record does not assert independent reproduction. ### Random Accuracy: 20. Result ID: 6992f1e53a6c314849128d1b. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.