# Pairwise LLM Judging on RewardBench (Coverage and Risk) Source: [Pairwise LLM Judging on RewardBench (Coverage and Risk)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:40:04.049Z Evaluation of LLM performance using pairwise comparison judging methodology on the RewardBench dataset, specifically reporting Coverage and Risk metrics. 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 | Pairwise LLM Judging | | Dataset (including any stated split) | RewardBench | | Metric | Coverage | | 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 - [Coverage](): Higher is better; unit: PERCENT. - [Risk](): Lower is better; unit: SCALAR. ## Recorded results | Method | Coverage | Result date | Report paper | | --- | --- | --- | --- | | Vanilla | 100 | Not recorded | 6992a4bc9ebc499c97406dba | | Vanilla | 100 | Not recorded | 6992a4bc9ebc499c97406dba | | Vanilla | 100 | Not recorded | 6992a4bc9ebc499c97406dba | | Vanilla | 100 | Not recorded | 6992a4bc9ebc499c97406dba | | SCOPE | 98.3 | Not recorded | 6992a4bc9ebc499c97406dba | | Heuristic | 96.8 | Not recorded | 6992a4bc9ebc499c97406dba | | Heuristic | 96.4 | Not recorded | 6992a4bc9ebc499c97406dba | | Naive | 95.7 | Not recorded | 6992a4bc9ebc499c97406dba | | SCOPE | 93.5 | Not recorded | 6992a4bc9ebc499c97406dba | | Heuristic | 89.4 | Not recorded | 6992a4bc9ebc499c97406dba | | SCOPE | 89.2 | Not recorded | 6992a4bc9ebc499c97406dba | | Naive | 87.2 | Not recorded | 6992a4bc9ebc499c97406dba | | Heuristic | 84.5 | Not recorded | 6992a4bc9ebc499c97406dba | | Naive | 82.7 | Not recorded | 6992a4bc9ebc499c97406dba | | SCOPE | 59.4 | Not recorded | 6992a4bc9ebc499c97406dba | | Naive | 43 | Not recorded | 6992a4bc9ebc499c97406dba | ### Vanilla Coverage: 100. Result ID: 699a62665698665ec4ea72fa. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-7B","alpha":"0.10"}. This record does not assert independent reproduction. ### Vanilla Coverage: 100. Result ID: 699a62665698665ec4ea72fe. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-14B","alpha":"0.10"}. This record does not assert independent reproduction. ### Vanilla Coverage: 100. Result ID: 699a62665698665ec4ea7302. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-32B","alpha":"0.10"}. This record does not assert independent reproduction. ### Vanilla Coverage: 100. Result ID: 699a62665698665ec4ea7306. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Llama-70B","alpha":"0.10"}. This record does not assert independent reproduction. ### SCOPE Coverage: 98.3. Result ID: 699a62665698665ec4ea7305. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-32B","alpha":"0.10"}. This record does not assert independent reproduction. ### Heuristic Coverage: 96.8. Result ID: 699a62665698665ec4ea72ff. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-14B","alpha":"0.10"}. This record does not assert independent reproduction. ### Heuristic Coverage: 96.4. Result ID: 699a62665698665ec4ea7303. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-32B","alpha":"0.10"}. This record does not assert independent reproduction. ### Naive Coverage: 95.7. Result ID: 699a62665698665ec4ea7304. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-32B","alpha":"0.10"}. This record does not assert independent reproduction. ### SCOPE Coverage: 93.5. Result ID: 699a62665698665ec4ea7309. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Judge Model":"Llama-70B","alpha":"0.10"}. This record does not assert independent reproduction. ### Heuristic Coverage: 89.4. Result ID: 699a62665698665ec4ea72fb. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-7B","alpha":"0.10"}. This record does not assert independent reproduction. ### SCOPE Coverage: 89.2. Result ID: 699a62665698665ec4ea7301. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-14B","alpha":"0.10"}. This record does not assert independent reproduction. ### Naive Coverage: 87.2. Result ID: 699a62665698665ec4ea7308. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Llama-70B","alpha":"0.10"}. This record does not assert independent reproduction. ### Heuristic Coverage: 84.5. Result ID: 699a62665698665ec4ea7307. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Llama-70B","alpha":"0.10"}. This record does not assert independent reproduction. ### Naive Coverage: 82.7. Result ID: 699a62665698665ec4ea7300. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-14B","alpha":"0.10"}. This record does not assert independent reproduction. ### SCOPE Coverage: 59.4. Result ID: 699a62665698665ec4ea72fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-7B","alpha":"0.10"}. This record does not assert independent reproduction. ### Naive Coverage: 43. Result ID: 699a62665698665ec4ea72fc. [Reporting paper]() Recorded evaluation configuration: {"Judge Model":"Qwen-7B","alpha":"0.10"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.