# Reward Aggregation on HealthBench 500 Source: [Reward Aggregation on HealthBench 500]() Collection index: [llms.txt]() Data updated: 2026-06-04T08:56:53.902Z Evaluation of model performance on the Reward Aggregation task using the HealthBench 500 dataset, measured by Score and Absolute Gain. 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 | Reward Aggregation | | Dataset (including any stated split) | HealthBench 500 | | Metric | 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 - [Score](): Higher is better; unit: SCALAR. - [Absolute Gain](): Higher is better; unit: SCALAR. ## Recorded results | Method | Score | Result date | Report paper | | --- | --- | --- | --- | | RuscaRL + GEAR | 65.7 | Not recorded | 6a210667c10484bac23eff5b | | RuscaRL + GEAR | 63.8 | Not recorded | 6a210667c10484bac23eff5b | | GEAR | 62.9 | Not recorded | 6a210667c10484bac23eff5b | | GEAR | 62.5 | Not recorded | 6a210667c10484bac23eff5b | | RuscaRL | 58.2 | Not recorded | 6a210667c10484bac23eff5b | | RuscaRL | 56.9 | Not recorded | 6a210667c10484bac23eff5b | | Flat aggregation | 56.7 | Not recorded | 6a210667c10484bac23eff5b | | RaR-Explicit + GEAR | 55 | Not recorded | 6a210667c10484bac23eff5b | | Hard gating | 54.6 | Not recorded | 6a210667c10484bac23eff5b | | Flat aggregation | 54.1 | Not recorded | 6a210667c10484bac23eff5b | | RaR-Explicit + GEAR | 54.1 | Not recorded | 6a210667c10484bac23eff5b | | Hard gating | 53 | Not recorded | 6a210667c10484bac23eff5b | | RaR-Explicit | 49.5 | Not recorded | 6a210667c10484bac23eff5b | | RaR-Explicit | 48.4 | Not recorded | 6a210667c10484bac23eff5b | | MeRF + GEAR | 45.6 | Not recorded | 6a210667c10484bac23eff5b | | MeRF + GEAR | 44.7 | Not recorded | 6a210667c10484bac23eff5b | | MeRF | 40.2 | Not recorded | 6a210667c10484bac23eff5b | | MeRF | 39.5 | Not recorded | 6a210667c10484bac23eff5b | ### RuscaRL + GEAR Score: 65.7. Result ID: 6a213dccfbc373592c1cf1c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Qwen2.5-7B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### RuscaRL + GEAR Score: 63.8. Result ID: 6a213dccfbc373592c1cf1c9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Llama-3.1-8B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### GEAR Score: 62.9. Result ID: 6a213dccfbc373592c1cf1ba. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Qwen2.5-7B-Instruct","Pipeline Category":"Aggregation-only"}. This record does not assert independent reproduction. ### GEAR Score: 62.5. Result ID: 6a213dccfbc373592c1cf1c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Llama-3.1-8B-Instruct","Pipeline Category":"Aggregation-only"}. This record does not assert independent reproduction. ### RuscaRL Score: 58.2. Result ID: 6a213dccfbc373592c1cf1bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Qwen2.5-7B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### RuscaRL Score: 56.9. Result ID: 6a213dccfbc373592c1cf1c8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Llama-3.1-8B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### Flat aggregation Score: 56.7. Result ID: 6a213dccfbc373592c1cf1b8. [Reporting paper]() Recorded evaluation configuration: {"Policy Model":"Qwen2.5-7B-Instruct","Pipeline Category":"Aggregation-only"}. This record does not assert independent reproduction. ### RaR-Explicit + GEAR Score: 55. Result ID: 6a213dccfbc373592c1cf1bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Qwen2.5-7B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### Hard gating Score: 54.6. Result ID: 6a213dccfbc373592c1cf1b9. [Reporting paper]() Recorded evaluation configuration: {"Policy Model":"Qwen2.5-7B-Instruct","Pipeline Category":"Aggregation-only"}. This record does not assert independent reproduction. ### Flat aggregation Score: 54.1. Result ID: 6a213dccfbc373592c1cf1c1. [Reporting paper]() Recorded evaluation configuration: {"Policy Model":"Llama-3.1-8B-Instruct","Pipeline Category":"Aggregation-only"}. This record does not assert independent reproduction. ### RaR-Explicit + GEAR Score: 54.1. Result ID: 6a213dccfbc373592c1cf1c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Llama-3.1-8B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### Hard gating Score: 53. Result ID: 6a213dccfbc373592c1cf1c2. [Reporting paper]() Recorded evaluation configuration: {"Policy Model":"Llama-3.1-8B-Instruct","Pipeline Category":"Aggregation-only"}. This record does not assert independent reproduction. ### RaR-Explicit Score: 49.5. Result ID: 6a213dccfbc373592c1cf1bb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Qwen2.5-7B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### RaR-Explicit Score: 48.4. Result ID: 6a213dccfbc373592c1cf1c4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Llama-3.1-8B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### MeRF + GEAR Score: 45.6. Result ID: 6a213dccfbc373592c1cf1be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Qwen2.5-7B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### MeRF + GEAR Score: 44.7. Result ID: 6a213dccfbc373592c1cf1c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Llama-3.1-8B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### MeRF Score: 40.2. Result ID: 6a213dccfbc373592c1cf1bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Qwen2.5-7B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. ### MeRF Score: 39.5. Result ID: 6a213dccfbc373592c1cf1c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Policy Model":"Llama-3.1-8B-Instruct","Pipeline Category":"Plug-in replacement in rubric-guided RL pipelines"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.