# LLM Steering on Persona Source: [LLM Steering on Persona]() Collection index: [llms.txt]() Data updated: 2026-05-29T10:53:48.829Z Evaluation of LLM steering performance using the Persona dataset, measured by average trait score. 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 | LLM Steering | | Dataset (including any stated split) | Persona | | Metric | Average Trait 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 - [Average Trait Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | Average Trait Score | Result date | Report paper | | --- | --- | --- | --- | | UNISTEER | 81.75 | Not recorded | 6a191de3dbd7150506f82bf4 | | CAA | 80.67 | Not recorded | 6a191de3dbd7150506f82bf4 | | UNISTEER | 77.67 | Not recorded | 6a191de3dbd7150506f82bf4 | | UNISTEER | 72.03 | Not recorded | 6a191de3dbd7150506f82bf4 | | CAA | 67.33 | Not recorded | 6a191de3dbd7150506f82bf4 | | LoReFT | 54.67 | Not recorded | 6a191de3dbd7150506f82bf4 | | ODESteer | 47.67 | Not recorded | 6a191de3dbd7150506f82bf4 | | LoReFT | 42.78 | Not recorded | 6a191de3dbd7150506f82bf4 | | RepE | 36.67 | Not recorded | 6a191de3dbd7150506f82bf4 | | RepE | 33.67 | Not recorded | 6a191de3dbd7150506f82bf4 | | RepE | 31.1 | Not recorded | 6a191de3dbd7150506f82bf4 | | CAA | 28 | Not recorded | 6a191de3dbd7150506f82bf4 | | LoReFT | 20.76 | Not recorded | 6a191de3dbd7150506f82bf4 | | Original | 19.37 | Not recorded | 6a191de3dbd7150506f82bf4 | | ODESteer | 16.6 | Not recorded | 6a191de3dbd7150506f82bf4 | | ODESteer | 11.67 | Not recorded | 6a191de3dbd7150506f82bf4 | | Original | 5.1 | Not recorded | 6a191de3dbd7150506f82bf4 | | Original | 4.1 | Not recorded | 6a191de3dbd7150506f82bf4 | ### UNISTEER Average Trait Score: 81.75. Result ID: 6a196fdcfbc373592c148c75. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### CAA Average Trait Score: 80.67. Result ID: 6a196fdcfbc373592c148c71. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### UNISTEER Average Trait Score: 77.67. Result ID: 6a196fdcfbc373592c148c6f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-1.5B"}. This record does not assert independent reproduction. ### UNISTEER Average Trait Score: 72.03. Result ID: 6a196fdcfbc373592c148c69. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Llama-3.2-1B"}. This record does not assert independent reproduction. ### CAA Average Trait Score: 67.33. Result ID: 6a196fdcfbc373592c148c65. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Llama-3.2-1B"}. This record does not assert independent reproduction. ### LoReFT Average Trait Score: 54.67. Result ID: 6a196fdcfbc373592c148c6d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-1.5B"}. This record does not assert independent reproduction. ### ODESteer Average Trait Score: 47.67. Result ID: 6a196fdcfbc373592c148c6e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-1.5B"}. This record does not assert independent reproduction. ### LoReFT Average Trait Score: 42.78. Result ID: 6a196fdcfbc373592c148c73. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### RepE Average Trait Score: 36.67. Result ID: 6a196fdcfbc373592c148c72. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### RepE Average Trait Score: 33.67. Result ID: 6a196fdcfbc373592c148c66. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Target LLM":"Llama-3.2-1B"}. This record does not assert independent reproduction. ### RepE Average Trait Score: 31.1. Result ID: 6a196fdcfbc373592c148c6c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-1.5B"}. This record does not assert independent reproduction. ### CAA Average Trait Score: 28. Result ID: 6a196fdcfbc373592c148c6b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-1.5B"}. This record does not assert independent reproduction. ### LoReFT Average Trait Score: 20.76. Result ID: 6a196fdcfbc373592c148c67. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Llama-3.2-1B"}. This record does not assert independent reproduction. ### Original Average Trait Score: 19.37. Result ID: 6a196fdcfbc373592c148c64. [Reporting paper]() Recorded evaluation configuration: {"Target LLM":"Llama-3.2-1B"}. This record does not assert independent reproduction. ### ODESteer Average Trait Score: 16.6. Result ID: 6a196fdcfbc373592c148c68. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Llama-3.2-1B"}. This record does not assert independent reproduction. ### ODESteer Average Trait Score: 11.67. Result ID: 6a196fdcfbc373592c148c74. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### Original Average Trait Score: 5.1. Result ID: 6a196fdcfbc373592c148c70. [Reporting paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### Original Average Trait Score: 4.1. Result ID: 6a196fdcfbc373592c148c6a. [Reporting paper]() Recorded evaluation configuration: {"Target LLM":"Qwen2.5-1.5B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.