# Latent Preference Modeling on MPT Context-Free, Preference Recall Source: [Latent Preference Modeling on MPT Context-Free, Preference Recall]() Collection index: [llms.txt]() Data updated: 2026-04-21T12:06:28.813Z Evaluation of model performance on the MPT Context-Free, Preference Recall dataset for the Latent Preference Modeling task, measured by Precision, Recall, and F1 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 | Latent Preference Modeling | | Dataset (including any stated split) | MPT Context-Free, Preference Recall | | Metric | Precision | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 19 of 19 | | 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 - [Precision](): Higher is better; unit: PERCENT. - [Recall](): Higher is better; unit: PERCENT. - [F1 Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Precision | Result date | Report paper | | --- | --- | --- | --- | | PREFINE | 76.1 | Not recorded | 69e7050b7d0172140ec7f237 | | PREFINE | 74.46 | Not recorded | 69e7050b7d0172140ec7f237 | | PREFINE | 73.23 | Not recorded | 69e7050b7d0172140ec7f237 | | PREFINE | 71.45 | Not recorded | 69e7050b7d0172140ec7f237 | | LangMem | 69.25 | Not recorded | 69e7050b7d0172140ec7f237 | | Gemini-3-Flash | 63.27 | Not recorded | 69e7050b7d0172140ec7f237 | | PREFINE | 62.11 | Not recorded | 69e7050b7d0172140ec7f237 | | GPT-5-mini | 61.42 | Not recorded | 69e7050b7d0172140ec7f237 | | GPT-5 | 59.39 | Not recorded | 69e7050b7d0172140ec7f237 | | RAG (Top-5) | 52.42 | Not recorded | 69e7050b7d0172140ec7f237 | | Mem0 | 52.36 | Not recorded | 69e7050b7d0172140ec7f237 | | GPT-4o-mini | 50.09 | Not recorded | 69e7050b7d0172140ec7f237 | | Gemma-3-12B | 47.78 | Not recorded | 69e7050b7d0172140ec7f237 | | PREFINE | 44.72 | Not recorded | 69e7050b7d0172140ec7f237 | | PREFINE | 36 | Not recorded | 69e7050b7d0172140ec7f237 | | PREFINE | 35.4 | Not recorded | 69e7050b7d0172140ec7f237 | | R1-Distill-Llama-8B | 32.29 | Not recorded | 69e7050b7d0172140ec7f237 | | R1-Distill-Qwen-7B | 21.12 | Not recorded | 69e7050b7d0172140ec7f237 | | CodeGemma-7B | 19.63 | Not recorded | 69e7050b7d0172140ec7f237 | ### PREFINE Precision: 76.1. Result ID: 69e7683947f34ce924cee49b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"Gemma-3-12B"}. This record does not assert independent reproduction. ### PREFINE Precision: 74.46. Result ID: 69e7683947f34ce924cee4a0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"GPT-5"}. This record does not assert independent reproduction. ### PREFINE Precision: 73.23. Result ID: 69e7683947f34ce924cee49f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"GPT-5-mini"}. This record does not assert independent reproduction. ### PREFINE Precision: 71.45. Result ID: 69e7683947f34ce924cee4a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"Gemini-3-Flash"}. This record does not assert independent reproduction. ### LangMem Precision: 69.25. Result ID: 69e7683947f34ce924cee499. [Reporting paper]() Recorded evaluation configuration: {"Methodology":"Memory-Augmented","Backbone":"Gemini-3-Flash"}. This record does not assert independent reproduction. ### Gemini-3-Flash Precision: 63.27. Result ID: 69e7683947f34ce924cee496. [Reporting paper]() Recorded evaluation configuration: {"Methodology":"Base Prompting"}. This record does not assert independent reproduction. ### PREFINE Precision: 62.11. Result ID: 69e7683947f34ce924cee49e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"GPT-4o-mini"}. This record does not assert independent reproduction. ### GPT-5-mini Precision: 61.42. Result ID: 69e7683947f34ce924cee494. [Reporting paper]() Recorded evaluation configuration: {"Methodology":"Base Prompting"}. This record does not assert independent reproduction. ### GPT-5 Precision: 59.39. Result ID: 69e7683947f34ce924cee495. [Reporting paper]() Recorded evaluation configuration: {"Methodology":"Base Prompting"}. This record does not assert independent reproduction. ### RAG (Top-5) Precision: 52.42. Result ID: 69e7683947f34ce924cee497. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Methodology":"Memory-Augmented","Backbone":"Gemini-3-Flash"}. This record does not assert independent reproduction. ### Mem0 Precision: 52.36. Result ID: 69e7683947f34ce924cee498. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Methodology":"Memory-Augmented","Backbone":"Gemini-3-Flash"}. This record does not assert independent reproduction. ### GPT-4o-mini Precision: 50.09. Result ID: 69e7683947f34ce924cee493. [Reporting paper]() Recorded evaluation configuration: {"Methodology":"Base Prompting"}. This record does not assert independent reproduction. ### Gemma-3-12B Precision: 47.78. Result ID: 69e7683947f34ce924cee490. [Reporting paper]() Recorded evaluation configuration: {"Methodology":"Base Prompting"}. This record does not assert independent reproduction. ### PREFINE Precision: 44.72. Result ID: 69e7683947f34ce924cee49c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"R1-Distill-Llama-8B"}. This record does not assert independent reproduction. ### PREFINE Precision: 36. Result ID: 69e7683947f34ce924cee49d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"R1-Distill-Qwen-7B"}. This record does not assert independent reproduction. ### PREFINE Precision: 35.4. Result ID: 69e7683947f34ce924cee49a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"CodeGemma-7B"}. This record does not assert independent reproduction. ### R1-Distill-Llama-8B Precision: 32.29. Result ID: 69e7683947f34ce924cee491. [Reporting paper]() Recorded evaluation configuration: {"Methodology":"Base Prompting"}. This record does not assert independent reproduction. ### R1-Distill-Qwen-7B Precision: 21.12. Result ID: 69e7683947f34ce924cee492. [Reporting paper]() Recorded evaluation configuration: {"Methodology":"Base Prompting"}. This record does not assert independent reproduction. ### CodeGemma-7B Precision: 19.63. Result ID: 69e7683947f34ce924cee48f. [Reporting paper]() Recorded evaluation configuration: {"Methodology":"Base Prompting"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.