# Best feasible prompt identification on XSum Source: [Best feasible prompt identification on XSum]() Collection index: [llms.txt]() Data updated: 2026-05-15T08:42:43.206Z Evaluation of the model's ability to identify the best feasible prompt using the XSum dataset, measured via soft constrained rewards. 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 | Best feasible prompt identification | | Dataset (including any stated split) | XSum | | Metric | Average Soft Constrained Reward (b=3) | | 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 - [Average Soft Constrained Reward (b=3)](): Higher is better; unit: SCALAR. - [Average Soft Constrained Reward (b=5)](): Higher is better; unit: SCALAR. - [Average Soft Constrained Reward (b=8)](): Higher is better; unit: SCALAR. - [Average Soft Constrained Reward (b=10)](): Higher is better; unit: SCALAR. ## Recorded results | Method | Average Soft Constrained Reward (b=3) | Result date | Report paper | | --- | --- | --- | --- | | CSR | 0.161 | Not recorded | 6a06a87cdbc836b87d829f83 | | CSR | 0.144 | Not recorded | 6a06a87cdbc836b87d829f83 | | MLP-CSR | 0.143 | Not recorded | 6a06a87cdbc836b87d829f83 | | MLP-CSR | 0.143 | Not recorded | 6a06a87cdbc836b87d829f83 | | MLP-CSR | 0.141 | Not recorded | 6a06a87cdbc836b87d829f83 | | CSR | 0.141 | Not recorded | 6a06a87cdbc836b87d829f83 | | MLP-CSR | 0.139 | Not recorded | 6a06a87cdbc836b87d829f83 | | CSR | 0.134 | Not recorded | 6a06a87cdbc836b87d829f83 | | MLP-CSR | 0.124 | Not recorded | 6a06a87cdbc836b87d829f83 | | MLP-CSR | 0.123 | Not recorded | 6a06a87cdbc836b87d829f83 | | CSR | 0.122 | Not recorded | 6a06a87cdbc836b87d829f83 | | CSR | 0.117 | Not recorded | 6a06a87cdbc836b87d829f83 | | Uniform | 0.113 | Not recorded | 6a06a87cdbc836b87d829f83 | | Uniform | 0.063 | Not recorded | 6a06a87cdbc836b87d829f83 | | Uniform | 0.048 | Not recorded | 6a06a87cdbc836b87d829f83 | | Uniform | 0.021 | Not recorded | 6a06a87cdbc836b87d829f83 | | Uniform | 0.021 | Not recorded | 6a06a87cdbc836b87d829f83 | | Uniform | 0.015 | Not recorded | 6a06a87cdbc836b87d829f83 | ### CSR Average Soft Constrained Reward (b=3): 0.161. Result ID: 6a06dc7447f34ce924ee9026. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama3-8B","K (Prompt pool size)":"30"}. This record does not assert independent reproduction. ### CSR Average Soft Constrained Reward (b=3): 0.144. Result ID: 6a06dc7447f34ce924ee9029. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama3-8B","K (Prompt pool size)":"50"}. This record does not assert independent reproduction. ### MLP-CSR Average Soft Constrained Reward (b=3): 0.143. Result ID: 6a06dc7447f34ce924ee9021. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Gemma-7B","K (Prompt pool size)":"50"}. This record does not assert independent reproduction. ### MLP-CSR Average Soft Constrained Reward (b=3): 0.143. Result ID: 6a06dc7447f34ce924ee9027. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama3-8B","K (Prompt pool size)":"30"}. This record does not assert independent reproduction. ### MLP-CSR Average Soft Constrained Reward (b=3): 0.141. Result ID: 6a06dc7447f34ce924ee902a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama3-8B","K (Prompt pool size)":"50"}. This record does not assert independent reproduction. ### CSR Average Soft Constrained Reward (b=3): 0.141. Result ID: 6a06dc7447f34ce924ee902c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama3-8B","K (Prompt pool size)":"100"}. This record does not assert independent reproduction. ### MLP-CSR Average Soft Constrained Reward (b=3): 0.139. Result ID: 6a06dc7447f34ce924ee9024. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Gemma-7B","K (Prompt pool size)":"100"}. This record does not assert independent reproduction. ### CSR Average Soft Constrained Reward (b=3): 0.134. Result ID: 6a06dc7447f34ce924ee9023. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Gemma-7B","K (Prompt pool size)":"100"}. This record does not assert independent reproduction. ### MLP-CSR Average Soft Constrained Reward (b=3): 0.124. Result ID: 6a06dc7447f34ce924ee902d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama3-8B","K (Prompt pool size)":"100"}. This record does not assert independent reproduction. ### MLP-CSR Average Soft Constrained Reward (b=3): 0.123. Result ID: 6a06dc7447f34ce924ee901e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Gemma-7B","K (Prompt pool size)":"30"}. This record does not assert independent reproduction. ### CSR Average Soft Constrained Reward (b=3): 0.122. Result ID: 6a06dc7447f34ce924ee9020. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Gemma-7B","K (Prompt pool size)":"50"}. This record does not assert independent reproduction. ### CSR Average Soft Constrained Reward (b=3): 0.117. Result ID: 6a06dc7447f34ce924ee901d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Gemma-7B","K (Prompt pool size)":"30"}. This record does not assert independent reproduction. ### Uniform Average Soft Constrained Reward (b=3): 0.113. Result ID: 6a06dc7447f34ce924ee902b. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Llama3-8B","K (Prompt pool size)":"100"}. This record does not assert independent reproduction. ### Uniform Average Soft Constrained Reward (b=3): 0.063. Result ID: 6a06dc7447f34ce924ee9028. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Llama3-8B","K (Prompt pool size)":"50"}. This record does not assert independent reproduction. ### Uniform Average Soft Constrained Reward (b=3): 0.048. Result ID: 6a06dc7447f34ce924ee9025. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Llama3-8B","K (Prompt pool size)":"30"}. This record does not assert independent reproduction. ### Uniform Average Soft Constrained Reward (b=3): 0.021. Result ID: 6a06dc7447f34ce924ee901f. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Gemma-7B","K (Prompt pool size)":"50"}. This record does not assert independent reproduction. ### Uniform Average Soft Constrained Reward (b=3): 0.021. Result ID: 6a06dc7447f34ce924ee9022. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Gemma-7B","K (Prompt pool size)":"100"}. This record does not assert independent reproduction. ### Uniform Average Soft Constrained Reward (b=3): 0.015. Result ID: 6a06dc7447f34ce924ee901c. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Gemma-7B","K (Prompt pool size)":"30"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.