# Text Classification on MR, CR, MPQA, SUBJ, TREC fine-tuning Noise Rate 40% Source: [Text Classification on MR, CR, MPQA, SUBJ, TREC fine-tuning Noise Rate 40%]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:50:17.404Z Evaluation of Text Classification performance on a composite dataset derived from MR, CR, MPQA, SUBJ, and TREC, specifically tested under a 40% noise rate condition during fine-tuning. 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 | Text Classification | | Dataset (including any stated split) | MR, CR, MPQA, SUBJ, TREC fine-tuning Noise Rate 40% | | Metric | MR Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 3 of 3 | | 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 - [MR Accuracy](): Higher is better; unit: PERCENT. - [CR Accuracy](): Higher is better; unit: PERCENT. - [MPQA Accuracy](): Higher is better; unit: PERCENT. - [SUBJ Accuracy](): Higher is better; unit: PERCENT. - [TREC Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | MR Accuracy | Result date | Report paper | | --- | --- | --- | --- | | REGSL | 83.54 | Not recorded | 68fa7a911e209131cea83f3d | | l2-PGM | 83.45 | Not recorded | 68fa7a911e209131cea83f3d | | Fine-tuning | 82.91 | Not recorded | 68fa7a911e209131cea83f3d | ### REGSL MR Accuracy: 83.54. Result ID: 698b79407a967eddb76ee355. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Architecture":"three-layer feed forward network","Pre-trained":"SST","Noise Rate":"40%","Number of random seeds":5}. This record does not assert independent reproduction. ### l2-PGM MR Accuracy: 83.45. Result ID: 698b79407a967eddb76ee354. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Architecture":"three-layer feed forward network","Pre-trained":"SST","Noise Rate":"40%","Number of random seeds":5}. This record does not assert independent reproduction. ### Fine-tuning MR Accuracy: 82.91. Result ID: 698b79407a967eddb76ee353. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"three-layer feed forward network","Pre-trained":"SST","Noise Rate":"40%","Number of random seeds":5}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.