# Question Answering on Pop-QA Cities 20 (Custom Metrics) Source: [Question Answering on Pop-QA Cities 20 (Custom Metrics)]() Collection index: [llms.txt]() Data updated: 2026-05-20T08:34:08.817Z Evaluation of Question Answering performance on the Pop-QA Cities 20 dataset, measured using custom quality metrics including Factual Accuracy, Completeness, Relevance, Clarity, and Binary Accuracy. 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 | Question Answering Evaluation | | Dataset (including any stated split) | Pop-QA Cities 20 | | Metric | Factual Accuracy | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [Factual Accuracy](): Higher is better; unit: SCALAR. - [Completeness](): Higher is better; unit: SCALAR. - [Relevance](): Higher is better; unit: SCALAR. - [Clarity](): Higher is better; unit: SCALAR. - [Binary Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Factual Accuracy | Result date | Report paper | | --- | --- | --- | --- | | InstructLab | 1.75 | Not recorded | 6a0d4045f984739ed67379f9 | | InstructLab | 1.73 | Not recorded | 6a0d4045f984739ed67379f9 | | EmbGen | 1.73 | Not recorded | 6a0d4045f984739ed67379f9 | | EmbGen | 1.68 | Not recorded | 6a0d4045f984739ed67379f9 | | Knowledge-Instruct | 1.54 | Not recorded | 6a0d4045f984739ed67379f9 | | Knowledge-Instruct | 1.54 | Not recorded | 6a0d4045f984739ed67379f9 | | EntiGraph | 1.39 | Not recorded | 6a0d4045f984739ed67379f9 | | EntiGraph | 1.35 | Not recorded | 6a0d4045f984739ed67379f9 | | Baseline: llama 8B Instruct | 1.25 | Not recorded | 6a0d4045f984739ed67379f9 | | Baseline: llama 8B Instruct | 1.25 | Not recorded | 6a0d4045f984739ed67379f9 | ### InstructLab Factual Accuracy: 1.75. Result ID: 6a0d71f5fbc373592c057d37. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"5M"}. This record does not assert independent reproduction. ### InstructLab Factual Accuracy: 1.73. Result ID: 6a0d71f5fbc373592c057d3c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"20M"}. This record does not assert independent reproduction. ### EmbGen Factual Accuracy: 1.73. Result ID: 6a0d71f5fbc373592c057d3e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"20M"}. This record does not assert independent reproduction. ### EmbGen Factual Accuracy: 1.68. Result ID: 6a0d71f5fbc373592c057d39. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"5M"}. This record does not assert independent reproduction. ### Knowledge-Instruct Factual Accuracy: 1.54. Result ID: 6a0d71f5fbc373592c057d38. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"5M"}. This record does not assert independent reproduction. ### Knowledge-Instruct Factual Accuracy: 1.54. Result ID: 6a0d71f5fbc373592c057d3d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"20M"}. This record does not assert independent reproduction. ### EntiGraph Factual Accuracy: 1.39. Result ID: 6a0d71f5fbc373592c057d36. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"5M"}. This record does not assert independent reproduction. ### EntiGraph Factual Accuracy: 1.35. Result ID: 6a0d71f5fbc373592c057d3b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"20M"}. This record does not assert independent reproduction. ### Baseline: llama 8B Instruct Factual Accuracy: 1.25. Result ID: 6a0d71f5fbc373592c057d35. [Reporting paper]() Recorded evaluation configuration: {"Size":"5M"}. This record does not assert independent reproduction. ### Baseline: llama 8B Instruct Factual Accuracy: 1.25. Result ID: 6a0d71f5fbc373592c057d3a. [Reporting paper]() Recorded evaluation configuration: {"Size":"20M"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.