# Language Understanding on In-domain (test) Source: [Language Understanding on In-domain (test)]() Collection index: [llms.txt]() Data updated: 2026-07-03T06:54:33.169Z Evaluation of general Language Understanding performance on the in-domain test set, reporting Accuracy, AUROC, and Expected Calibration Error (ECE). 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 | Language Understanding | | Dataset (including any stated split) | In-domain (test) | | Metric | Accuracy | | 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 - [Accuracy](): Higher is better; unit: PERCENT. - [AUROC](): Higher is better; unit: SCALAR. - [ECE](): Lower is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | RLVR | 66.6 | Not recorded | 6a474208143ebdcacacc5bf1 | | RLCR | 65.1 | Not recorded | 6a474208143ebdcacacc5bf1 | | C3RL w/o Ref | 64.9 | Not recorded | 6a474208143ebdcacacc5bf1 | | C3RL | 64.4 | Not recorded | 6a474208143ebdcacacc5bf1 | | SaySelf | 55 | Not recorded | 6a474208143ebdcacacc5bf1 | | SC | 53.9 | Not recorded | 6a474208143ebdcacacc5bf1 | | C3RL | 49.1 | Not recorded | 6a474208143ebdcacacc5bf1 | | SFT+Ref | 48 | Not recorded | 6a474208143ebdcacacc5bf1 | | C3RL w/o Ref | 47.3 | Not recorded | 6a474208143ebdcacacc5bf1 | | RLVR | 45.9 | Not recorded | 6a474208143ebdcacacc5bf1 | | Base | 45.3 | Not recorded | 6a474208143ebdcacacc5bf1 | | RLCR | 43.3 | Not recorded | 6a474208143ebdcacacc5bf1 | | SC | 40.8 | Not recorded | 6a474208143ebdcacacc5bf1 | | SFT+Ref | 37.6 | Not recorded | 6a474208143ebdcacacc5bf1 | | Base | 37.2 | Not recorded | 6a474208143ebdcacacc5bf1 | | SaySelf | 13.1 | Not recorded | 6a474208143ebdcacacc5bf1 | ### RLVR Accuracy: 66.6. Result ID: 6a475c8a60b88bb339923e2a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen2.5VL-7B-Instruct"}. This record does not assert independent reproduction. ### RLCR Accuracy: 65.1. Result ID: 6a475c8a60b88bb339923e2c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen2.5VL-7B-Instruct"}. This record does not assert independent reproduction. ### C3RL w/o Ref Accuracy: 64.9. Result ID: 6a475c8a60b88bb339923e2d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen2.5VL-7B-Instruct"}. This record does not assert independent reproduction. ### C3RL Accuracy: 64.4. Result ID: 6a475c8a60b88bb339923e2e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen2.5VL-7B-Instruct"}. This record does not assert independent reproduction. ### SaySelf Accuracy: 55. Result ID: 6a475c8a60b88bb339923e2b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen2.5VL-7B-Instruct"}. This record does not assert independent reproduction. ### SC Accuracy: 53.9. Result ID: 6a475c8a60b88bb339923e28. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen2.5VL-7B-Instruct"}. This record does not assert independent reproduction. ### C3RL Accuracy: 49.1. Result ID: 6a475c8a60b88bb339923e36. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.2-3B-Instruct"}. This record does not assert independent reproduction. ### SFT+Ref Accuracy: 48. Result ID: 6a475c8a60b88bb339923e29. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen2.5VL-7B-Instruct"}. This record does not assert independent reproduction. ### C3RL w/o Ref Accuracy: 47.3. Result ID: 6a475c8a60b88bb339923e35. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.2-3B-Instruct"}. This record does not assert independent reproduction. ### RLVR Accuracy: 45.9. Result ID: 6a475c8a60b88bb339923e32. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.2-3B-Instruct"}. This record does not assert independent reproduction. ### Base Accuracy: 45.3. Result ID: 6a475c8a60b88bb339923e27. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen2.5VL-7B-Instruct"}. This record does not assert independent reproduction. ### RLCR Accuracy: 43.3. Result ID: 6a475c8a60b88bb339923e34. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.2-3B-Instruct"}. This record does not assert independent reproduction. ### SC Accuracy: 40.8. Result ID: 6a475c8a60b88bb339923e30. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.2-3B-Instruct"}. This record does not assert independent reproduction. ### SFT+Ref Accuracy: 37.6. Result ID: 6a475c8a60b88bb339923e31. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.2-3B-Instruct"}. This record does not assert independent reproduction. ### Base Accuracy: 37.2. Result ID: 6a475c8a60b88bb339923e2f. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.2-3B-Instruct"}. This record does not assert independent reproduction. ### SaySelf Accuracy: 13.1. Result ID: 6a475c8a60b88bb339923e33. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.2-3B-Instruct"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.