# Multiple-choice reading comprehension on ViMMRC 2.0 Source: [Multiple-choice reading comprehension on ViMMRC 2.0]() Collection index: [llms.txt]() Data updated: 2026-03-24T14:44:52.096Z Evaluation of model performance on the ViMMRC 2.0 dataset for multiple-choice reading comprehension tasks, measured using Accuracy and F1-macro. 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 | Multiple-choice reading comprehension | | Dataset (including any stated split) | ViMMRC 2.0 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 29 of 29 | | 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. - [F1-macro](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | ViMultiChoice | 60.32 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | ViCLSR | 59.06 | Not recorded | 69c21ab90636e5bf4afc4b18 | | viBERT | 58.81 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | CafeBERT | 57.98 | Not recorded | 69c21ab90636e5bf4afc4b18 | | XLM-R | 57.52 | Not recorded | 69c21ab90636e5bf4afc4b18 | | viBERT | 57.17 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | CTM | 57.17 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | mBERT | 56.99 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | viBERT | 56.81 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | ViT5 Encoder | 56.63 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | OCN | 56.27 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | DCMN+ | 55.73 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | mBERT | 55.64 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | PhoBERT | 54.73 | Not recorded | 69c21ab90636e5bf4afc4b18 | | PhoBERT | 53.92 | Not recorded | 69c21ab90636e5bf4afc4b18 | | viBERT | 53.74 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | mBERT | 53.38 | Not recorded | 69c21ab90636e5bf4afc4b18 | | mBERT | 52.93 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | MMN | 52.93 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | XLM-R | 51.84 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | mBERT | 47.79 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | DynSAN | 44.36 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | BERT4News | 41.74 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | CoMatch | 41.21 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | DiffCSE | 41.11 | Not recorded | 69c21ab90636e5bf4afc4b18 | | ElimiNET | 38.59 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | CNN for MCQA | 35.08 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | HAF | 31.65 | Not recorded | 698c0d2bc97a0b3e3b5e123e | | XLM-R | 29.39 | Not recorded | 69c21ab90636e5bf4afc4b18 | ### ViMultiChoice Accuracy: 60.32. Result ID: 69963c635698665ec47f6293. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Explanation Generator":"w/o"}. This record does not assert independent reproduction. ### ViCLSR Accuracy: 59.06. Result ID: 69c2a35747f34ce924ab6430. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### viBERT Accuracy: 58.81. Result ID: 69963c635698665ec47f6287. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Multi-step Attention Network":"true","Natural Language Inference":"true"}. This record does not assert independent reproduction. ### CafeBERT Accuracy: 57.98. Result ID: 69c2a35747f34ce924ab642e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### XLM-R Accuracy: 57.52. Result ID: 69c2a35747f34ce924ab642d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"variant":"Large"}. This record does not assert independent reproduction. ### viBERT Accuracy: 57.17. Result ID: 69963c635698665ec47f6285. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Multi-step Attention Network":"true"}. This record does not assert independent reproduction. ### CTM Accuracy: 57.17. Result ID: 69963c635698665ec47f628f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### mBERT Accuracy: 56.99. Result ID: 69963c635698665ec47f6281. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Natural Language Inference":"true"}. This record does not assert independent reproduction. ### viBERT Accuracy: 56.81. Result ID: 69963c635698665ec47f6286. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Natural Language Inference":"true"}. This record does not assert independent reproduction. ### ViT5 Encoder Accuracy: 56.63. Result ID: 69963c635698665ec47f6291. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### OCN Accuracy: 56.27. Result ID: 69963c635698665ec47f6292. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DCMN+ Accuracy: 55.73. Result ID: 69963c635698665ec47f6290. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### mBERT Accuracy: 55.64. Result ID: 69963c635698665ec47f6282. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Multi-step Attention Network":"true","Natural Language Inference":"true"}. This record does not assert independent reproduction. ### PhoBERT Accuracy: 54.73. Result ID: 69c2a35747f34ce924ab642a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"variant":"Large"}. This record does not assert independent reproduction. ### PhoBERT Accuracy: 53.92. Result ID: 69c2a35747f34ce924ab6429. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"variant":"Base"}. This record does not assert independent reproduction. ### viBERT Accuracy: 53.74. Result ID: 69963c635698665ec47f6284. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### mBERT Accuracy: 53.38. Result ID: 69c2a35747f34ce924ab642b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### mBERT Accuracy: 52.93. Result ID: 69963c635698665ec47f6280. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Multi-step Attention Network":"true"}. This record does not assert independent reproduction. ### MMN Accuracy: 52.93. Result ID: 69963c635698665ec47f628d. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### XLM-R Accuracy: 51.84. Result ID: 69963c635698665ec47f6283. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### mBERT Accuracy: 47.79. Result ID: 69963c635698665ec47f627f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DynSAN Accuracy: 44.36. Result ID: 69963c635698665ec47f628e. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BERT4News Accuracy: 41.74. Result ID: 69963c635698665ec47f6288. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### CoMatch Accuracy: 41.21. Result ID: 69963c635698665ec47f628b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DiffCSE Accuracy: 41.11. Result ID: 69c2a35747f34ce924ab642f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ElimiNET Accuracy: 38.59. Result ID: 69963c635698665ec47f628c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### CNN for MCQA Accuracy: 35.08. Result ID: 69963c635698665ec47f628a. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### HAF Accuracy: 31.65. Result ID: 69963c635698665ec47f6289. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### XLM-R Accuracy: 29.39. Result ID: 69c2a35747f34ce924ab642c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"variant":"Base"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.