# Knowledge Elicitation on PARAREL (test) Source: [Knowledge Elicitation on PARAREL (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:05:02.393Z Evaluation of model performance on the test split of the PARAREL dataset for Knowledge Elicitation tasks, measured by success rate. 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 | Knowledge Elicitation | | Dataset (including any stated split) | PARAREL (test) | | Metric | Success Rate | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 28 of 28 | | 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 - [Success Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Success Rate | Result date | Report paper | | --- | --- | --- | --- | | PGDC | 71.36 | Not recorded | 68fa73681e209131cea5c4b3 | | Few-shots | 69.69 | Not recorded | 68fa73681e209131cea5c4b3 | | PGDC | 69.63 | Not recorded | 68fa73681e209131cea5c4b3 | | Few-shots | 66.95 | Not recorded | 68fa73681e209131cea5c4b3 | | Few-shots | 59.56 | Not recorded | 68fa73681e209131cea5c4b3 | | Few-shots | 58.23 | Not recorded | 68fa73681e209131cea5c4b3 | | PGDC | 55.95 | Not recorded | 68fa73681e209131cea5c4b3 | | Zero-shot | 54.97 | Not recorded | 68fa73681e209131cea5c4b3 | | Zero-shot | 54.78 | Not recorded | 68fa73681e209131cea5c4b3 | | Few-shots | 54.06 | Not recorded | 68fa73681e209131cea5c4b3 | | PGDC | 47.68 | Not recorded | 68fa73681e209131cea5c4b3 | | Few-shots | 44.84 | Not recorded | 68fa73681e209131cea5c4b3 | | Discriminator | 44.16 | Not recorded | 68fa73681e209131cea5c4b3 | | Zero-shot | 40.78 | Not recorded | 68fa73681e209131cea5c4b3 | | Zero-shot | 34.43 | Not recorded | 68fa73681e209131cea5c4b3 | | Zero-shot | 34.19 | Not recorded | 68fa73681e209131cea5c4b3 | | Zero-shot | 23.23 | Not recorded | 68fa73681e209131cea5c4b3 | | Discriminator | 23.06 | Not recorded | 68fa73681e209131cea5c4b3 | | Few-shots | 20.46 | Not recorded | 68fa73681e209131cea5c4b3 | | Zero-shot | 18.18 | Not recorded | 68fa73681e209131cea5c4b3 | | Discriminator | 17.96 | Not recorded | 68fa73681e209131cea5c4b3 | | Few-shots | 12.61 | Not recorded | 68fa73681e209131cea5c4b3 | | Zero-shot | 9.27 | Not recorded | 68fa73681e209131cea5c4b3 | | Discriminator | 9.01 | Not recorded | 68fa73681e209131cea5c4b3 | | Discriminator | 8.4 | Not recorded | 68fa73681e209131cea5c4b3 | | Discriminator | 7.25 | Not recorded | 68fa73681e209131cea5c4b3 | | Discriminator | 3.13 | Not recorded | 68fa73681e209131cea5c4b3 | | Discriminator | 2.4 | Not recorded | 68fa73681e209131cea5c4b3 | ### PGDC Success Rate: 71.36. Result ID: 6992c8773a6c314849117a73. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA2 (7B)"}. This record does not assert independent reproduction. ### Few-shots Success Rate: 69.69. Result ID: 6992c8773a6c314849117a78. [Reporting paper]() Recorded evaluation configuration: {"Model":"Vicuna (7B)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### PGDC Success Rate: 69.63. Result ID: 6992c8773a6c314849117a7a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Vicuna (7B)"}. This record does not assert independent reproduction. ### Few-shots Success Rate: 66.95. Result ID: 6992c8773a6c314849117a71. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA2 (7B)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Few-shots Success Rate: 59.56. Result ID: 6992c8773a6c314849117a75. [Reporting paper]() Recorded evaluation configuration: {"Model":"Vicuna (7B)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Few-shots Success Rate: 58.23. Result ID: 6992c8773a6c314849117a6e. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA2 (7B)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### PGDC Success Rate: 55.95. Result ID: 6992c8773a6c314849117a81. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J (6B)"}. This record does not assert independent reproduction. ### Zero-shot Success Rate: 54.97. Result ID: 6992c8773a6c314849117a77. [Reporting paper]() Recorded evaluation configuration: {"Model":"Vicuna (7B)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Zero-shot Success Rate: 54.78. Result ID: 6992c8773a6c314849117a70. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA2 (7B)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Few-shots Success Rate: 54.06. Result ID: 6992c8773a6c314849117a7f. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-J (6B)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### PGDC Success Rate: 47.68. Result ID: 6992c8773a6c314849117a88. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-2 (774M)"}. This record does not assert independent reproduction. ### Few-shots Success Rate: 44.84. Result ID: 6992c8773a6c314849117a7c. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-J (6B)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Discriminator Success Rate: 44.16. Result ID: 6992c8773a6c314849117a72. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA2 (7B)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Zero-shot Success Rate: 40.78. Result ID: 6992c8773a6c314849117a7e. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-J (6B)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Zero-shot Success Rate: 34.43. Result ID: 6992c8773a6c314849117a6d. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA2 (7B)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Zero-shot Success Rate: 34.19. Result ID: 6992c8773a6c314849117a74. [Reporting paper]() Recorded evaluation configuration: {"Model":"Vicuna (7B)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Zero-shot Success Rate: 23.23. Result ID: 6992c8773a6c314849117a7b. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-J (6B)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Discriminator Success Rate: 23.06. Result ID: 6992c8773a6c314849117a79. [Reporting paper]() Recorded evaluation configuration: {"Model":"Vicuna (7B)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Few-shots Success Rate: 20.46. Result ID: 6992c8773a6c314849117a86. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-2 (774M)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Zero-shot Success Rate: 18.18. Result ID: 6992c8773a6c314849117a85. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-2 (774M)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Discriminator Success Rate: 17.96. Result ID: 6992c8773a6c314849117a6f. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA2 (7B)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Few-shots Success Rate: 12.61. Result ID: 6992c8773a6c314849117a83. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-2 (774M)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Zero-shot Success Rate: 9.27. Result ID: 6992c8773a6c314849117a82. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-2 (774M)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Discriminator Success Rate: 9.01. Result ID: 6992c8773a6c314849117a87. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-2 (774M)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Discriminator Success Rate: 8.4. Result ID: 6992c8773a6c314849117a76. [Reporting paper]() Recorded evaluation configuration: {"Model":"Vicuna (7B)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Discriminator Success Rate: 7.25. Result ID: 6992c8773a6c314849117a80. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-J (6B)","Prompt Variant":"paraphrased"}. This record does not assert independent reproduction. ### Discriminator Success Rate: 3.13. Result ID: 6992c8773a6c314849117a84. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-2 (774M)","Prompt Variant":"standard"}. This record does not assert independent reproduction. ### Discriminator Success Rate: 2.4. Result ID: 6992c8773a6c314849117a7d. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-J (6B)","Prompt Variant":"standard"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.