# Reading Comprehension on Race M (Score) Source: [Reading Comprehension on Race M (Score)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:00:55.906Z Evaluation of Reading Comprehension performance on the Race M (Medium) subset of the RACE dataset, measured by the reported 'Score'. 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 | Reading Comprehension | | Dataset (including any stated split) | Race M | | Metric | Race M Score | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [Race M Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | Race M Score | Result date | Report paper | | --- | --- | --- | --- | | RWKV-7B | 45.47 | Not recorded | 68fa75391e209131cea668f7 | | ReplaceMe (LS) | 37.1 | Not recorded | 68fa7be51e209131cea8b22c | | ReplaceMe (Cosine) | 35.4 | Not recorded | 68fa7be51e209131cea8b22c | | UIDL | 35.2 | Not recorded | 68fa7be51e209131cea8b22c | | LLM-Streamline | 34.8 | Not recorded | 68fa7be51e209131cea8b22c | | Llama 2 7B (baseline) | 33.1 | Not recorded | 68fa7be51e209131cea8b22c | | SliceGPT | 27 | Not recorded | 68fa7be51e209131cea8b22c | | ShortGPT | 25.77 | Not recorded | 68fa75391e209131cea668f7 | | Mamba-2.8B | 25.3 | Not recorded | 68fa75391e209131cea668f7 | | ShortGPT | 25.21 | Not recorded | 68fa75391e209131cea668f7 | | ShortGPT | 23.75 | Not recorded | 68fa75391e209131cea668f7 | | ShortGPT | 23.61 | Not recorded | 68fa75391e209131cea668f7 | | LaCo | 23.6 | Not recorded | 68fa7be51e209131cea8b22c | | ShortGPT | 23.26 | Not recorded | 68fa75391e209131cea668f7 | | LLMPruner | 22.9 | Not recorded | 68fa7be51e209131cea8b22c | | ShortGPT | 22.77 | Not recorded | 68fa75391e209131cea668f7 | | ShortGPT | 21.59 | Not recorded | 68fa75391e209131cea668f7 | | ShortGPT | 21.31 | Not recorded | 68fa75391e209131cea668f7 | ### RWKV-7B Race M Score: 45.47. Result ID: 699381123a6c314849166d01. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"RWKV-7B","Pruning ratio":"0%"}. This record does not assert independent reproduction. ### ReplaceMe (LS) Race M Score: 37.1. Result ID: 699b0d215698665ec4a02901. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Train-Free":true,"Base Model":"Llama 2 7B","Compression Ratio":"25%"}. This record does not assert independent reproduction. ### ReplaceMe (Cosine) Race M Score: 35.4. Result ID: 699b0d215698665ec4a02900. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Train-Free":true,"Base Model":"Llama 2 7B","Compression Ratio":"25%"}. This record does not assert independent reproduction. ### UIDL Race M Score: 35.2. Result ID: 699b0d215698665ec4a028ff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Train-Free":false,"Base Model":"Llama 2 7B","Compression Ratio":"25%"}. This record does not assert independent reproduction. ### LLM-Streamline Race M Score: 34.8. Result ID: 699b0d215698665ec4a028fb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Train-Free":false,"Base Model":"Llama 2 7B","Compression Ratio":"25%"}. This record does not assert independent reproduction. ### Llama 2 7B (baseline) Race M Score: 33.1. Result ID: 699b0d215698665ec4a028fa. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama 2 7B","Compression Ratio":"0%"}. This record does not assert independent reproduction. ### SliceGPT Race M Score: 27. Result ID: 699b0d215698665ec4a028fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Train-Free":false,"Base Model":"Llama 2 7B","Compression Ratio":"25%"}. This record does not assert independent reproduction. ### ShortGPT Race M Score: 25.77. Result ID: 699381123a6c314849166d02. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RWKV-7B","Pruning ratio":"9.4%"}. This record does not assert independent reproduction. ### Mamba-2.8B Race M Score: 25.3. Result ID: 699381123a6c314849166cfc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mamba-2.8B","Pruning ratio":"0%"}. This record does not assert independent reproduction. ### ShortGPT Race M Score: 25.21. Result ID: 699381123a6c314849166cfd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mamba-2.8B","Pruning ratio":"10.9%"}. This record does not assert independent reproduction. ### ShortGPT Race M Score: 23.75. Result ID: 699381123a6c314849166d03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RWKV-7B","Pruning ratio":"18.8%"}. This record does not assert independent reproduction. ### ShortGPT Race M Score: 23.61. Result ID: 699381123a6c314849166cfe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mamba-2.8B","Pruning ratio":"20.3%"}. This record does not assert independent reproduction. ### LaCo Race M Score: 23.6. Result ID: 699b0d215698665ec4a028fe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Train-Free":false,"Base Model":"Llama 2 7B","Compression Ratio":"25%"}. This record does not assert independent reproduction. ### ShortGPT Race M Score: 23.26. Result ID: 699381123a6c314849166cff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mamba-2.8B","Pruning ratio":"25%"}. This record does not assert independent reproduction. ### LLMPruner Race M Score: 22.9. Result ID: 699b0d215698665ec4a028fc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Train-Free":false,"Base Model":"Llama 2 7B","Compression Ratio":"25%"}. This record does not assert independent reproduction. ### ShortGPT Race M Score: 22.77. Result ID: 699381123a6c314849166d00. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mamba-2.8B","Pruning ratio":"31.3%"}. This record does not assert independent reproduction. ### ShortGPT Race M Score: 21.59. Result ID: 699381123a6c314849166d04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RWKV-7B","Pruning ratio":"25%"}. This record does not assert independent reproduction. ### ShortGPT Race M Score: 21.31. Result ID: 699381123a6c314849166d05. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RWKV-7B","Pruning ratio":"28.1%"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.