# Personalized Agentic Social Support on ComPASS-Bench History-based setting Source: [Personalized Agentic Social Support on ComPASS-Bench History-based setting]() Collection index: [llms.txt]() Data updated: 2026-04-21T13:20:51.357Z Evaluation of personalized agentic social support capabilities using the ComPASS-Bench dataset, specifically focusing on history-based settings and measured by Preference Alignment Scores. 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 | Personalized Agentic Social Support | | Dataset (including any stated split) | ComPASS-Bench History-based setting | | Metric | Preference Alignment Score (w/o History) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Preference Alignment Score (w/o History)](): Higher is better; unit: SCALAR. - [Preference Alignment Score (w/ History)](): Higher is better; unit: SCALAR. - [Score Delta (w/ His. - w/o His.)](): Higher is better; unit: SCALAR. ## Recorded results | Method | Preference Alignment Score (w/o History) | Result date | Report paper | | --- | --- | --- | --- | | GPT-5.1 | 2.83 | Not recorded | 69e705247d0172140ec7f2e1 | | GPT-5.1 | 2.77 | Not recorded | 69e705247d0172140ec7f2e1 | | Qwen3-32B | 2.6 | Not recorded | 69e705247d0172140ec7f2e1 | | Gemini-3-Pro | 2.46 | Not recorded | 69e705247d0172140ec7f2e1 | | Qwen3-Max | 2.46 | Not recorded | 69e705247d0172140ec7f2e1 | | Qwen3-32B | 2.39 | Not recorded | 69e705247d0172140ec7f2e1 | | Qwen3-8B | 2.37 | Not recorded | 69e705247d0172140ec7f2e1 | | Claude-Sonnet-4.5 | 2.36 | Not recorded | 69e705247d0172140ec7f2e1 | | DeepSeek-V3.2 | 2.36 | Not recorded | 69e705247d0172140ec7f2e1 | | ComPASS-Qwen | 2.36 | Not recorded | 69e705247d0172140ec7f2e1 | | Qwen3-8B | 2.32 | Not recorded | 69e705247d0172140ec7f2e1 | | Llama-3.1-8B | 1.93 | Not recorded | 69e705247d0172140ec7f2e1 | ### GPT-5.1 Preference Alignment Score (w/o History): 2.83. Result ID: 69e7799b47f34ce924cf58cf. [Reporting paper]() Recorded evaluation configuration: {"Mode":"Instant"}. This record does not assert independent reproduction. ### GPT-5.1 Preference Alignment Score (w/o History): 2.77. Result ID: 69e7799b47f34ce924cf58d8. [Reporting paper]() Recorded evaluation configuration: {"Mode":"Reasoning"}. This record does not assert independent reproduction. ### Qwen3-32B Preference Alignment Score (w/o History): 2.6. Result ID: 69e7799b47f34ce924cf58d9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mode":"Reasoning"}. This record does not assert independent reproduction. ### Gemini-3-Pro Preference Alignment Score (w/o History): 2.46. Result ID: 69e7799b47f34ce924cf58d0. [Reporting paper]() Recorded evaluation configuration: {"Mode":"Instant"}. This record does not assert independent reproduction. ### Qwen3-Max Preference Alignment Score (w/o History): 2.46. Result ID: 69e7799b47f34ce924cf58d3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mode":"Instant"}. This record does not assert independent reproduction. ### Qwen3-32B Preference Alignment Score (w/o History): 2.39. Result ID: 69e7799b47f34ce924cf58d4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mode":"Instant"}. This record does not assert independent reproduction. ### Qwen3-8B Preference Alignment Score (w/o History): 2.37. Result ID: 69e7799b47f34ce924cf58da. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mode":"Reasoning"}. This record does not assert independent reproduction. ### Claude-Sonnet-4.5 Preference Alignment Score (w/o History): 2.36. Result ID: 69e7799b47f34ce924cf58d1. [Reporting paper]() Recorded evaluation configuration: {"Mode":"Instant"}. This record does not assert independent reproduction. ### DeepSeek-V3.2 Preference Alignment Score (w/o History): 2.36. Result ID: 69e7799b47f34ce924cf58d2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mode":"Instant"}. This record does not assert independent reproduction. ### ComPASS-Qwen Preference Alignment Score (w/o History): 2.36. Result ID: 69e7799b47f34ce924cf58d7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mode":"Instant"}. This record does not assert independent reproduction. ### Qwen3-8B Preference Alignment Score (w/o History): 2.32. Result ID: 69e7799b47f34ce924cf58d5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mode":"Instant"}. This record does not assert independent reproduction. ### Llama-3.1-8B Preference Alignment Score (w/o History): 1.93. Result ID: 69e7799b47f34ce924cf58d6. [Reporting paper]() Recorded evaluation configuration: {"Mode":"Instant"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.