# Graph Question Answering on GRBENCH Healthcare 1.0 (test) Source: [Graph Question Answering on GRBENCH Healthcare 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:59:13.990Z Evaluation of model performance on the test split of the GRBENCH Healthcare 1.0 dataset for Graph Question Answering, measured using R-L and GPT4score. 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 | Graph Question Answering | | Dataset (including any stated split) | GRBENCH Healthcare 1.0 (test) | | Metric | R-L Score | | 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 - [R-L Score](): Higher is better; unit: SCALAR. - [GPT-4 Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | R-L Score | Result date | Report paper | | --- | --- | --- | --- | | GRAPH-COT | 22.33 | Not recorded | 68fa751b1e209131cea65d00 | | GPT-3.5-turbo | 8.69 | Not recorded | 68fa751b1e209131cea65d00 | | Mixtral-8x7b | 8.66 | Not recorded | 68fa751b1e209131cea65d00 | | GPT-3.5-turbo | 5.83 | Not recorded | 68fa751b1e209131cea65d00 | | LLaMA-2-13b-chat | 5.25 | Not recorded | 68fa751b1e209131cea65d00 | | GPT-3.5-turbo | 4.57 | Not recorded | 68fa751b1e209131cea65d00 | | Mixtral-8x7b | 3.88 | Not recorded | 68fa751b1e209131cea65d00 | | Mixtral-8x7b | 3.22 | Not recorded | 68fa751b1e209131cea65d00 | | LLaMA-2-13b | 2.97 | Not recorded | 68fa751b1e209131cea65d00 | | LLaMA-2-13b-chat | 1.44 | Not recorded | 68fa751b1e209131cea65d00 | ### GRAPH-COT R-L Score: 22.33. Result ID: 699374ae3a6c3148491610da. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"GPT-3.5-turbo-16k"}. This record does not assert independent reproduction. ### GPT-3.5-turbo R-L Score: 8.69. Result ID: 699374ae3a6c3148491610d9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Graph RAG"}. This record does not assert independent reproduction. ### Mixtral-8x7b R-L Score: 8.66. Result ID: 699374ae3a6c3148491610d8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Graph RAG"}. This record does not assert independent reproduction. ### GPT-3.5-turbo R-L Score: 5.83. Result ID: 699374ae3a6c3148491610d3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Base"}. This record does not assert independent reproduction. ### LLaMA-2-13b-chat R-L Score: 5.25. Result ID: 699374ae3a6c3148491610d1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Base"}. This record does not assert independent reproduction. ### GPT-3.5-turbo R-L Score: 4.57. Result ID: 699374ae3a6c3148491610d6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Text RAG"}. This record does not assert independent reproduction. ### Mixtral-8x7b R-L Score: 3.88. Result ID: 699374ae3a6c3148491610d2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Base"}. This record does not assert independent reproduction. ### Mixtral-8x7b R-L Score: 3.22. Result ID: 699374ae3a6c3148491610d5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Text RAG"}. This record does not assert independent reproduction. ### LLaMA-2-13b R-L Score: 2.97. Result ID: 699374ae3a6c3148491610d7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Graph RAG"}. This record does not assert independent reproduction. ### LLaMA-2-13b-chat R-L Score: 1.44. Result ID: 699374ae3a6c3148491610d4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Text RAG"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.