# PRP Faithfulness Evaluation on Bob original characters 19 persona statements Source: [PRP Faithfulness Evaluation on Bob original characters 19 persona statements]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:11:24.490Z Evaluation of PRP Faithfulness performance on the Bob original characters 19 persona statements dataset, measured using ΔAPC metrics derived from DeB, GPT-4, and Human baselines. 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 | PRP Faithfulness Evaluation | | Dataset (including any stated split) | Bob original characters 19 persona statements | | Metric | ΔAPC (DeB) | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 14 of 14 | | 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 - [ΔAPC (DeB)](): Lower is better; unit: SCALAR. - [ΔAPC (GPT-4)](): Lower is better; unit: SCALAR. - [Human Reference Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | ΔAPC (DeB) | Result date | Report paper | | --- | --- | --- | --- | | Gemma-7B | 1.1 | Not recorded | 68fa753e1e209131cea66b68 | | Experience Upload (EU) | 3.4 | Not recorded | 68fa753e1e209131cea66b68 | | Experience Upload (EU) | 3.8 | Not recorded | 68fa753e1e209131cea66b68 | | Retrieval-augmented Generation (RAG) | 4 | Not recorded | 68fa753e1e209131cea66b68 | | Long-context Memory (LCM) | 4.5 | Not recorded | 68fa753e1e209131cea66b68 | | Retrieval-augmented Generation (RAG) | 5.2 | Not recorded | 68fa753e1e209131cea66b68 | | Long-context Memory (LCM) | 5.3 | Not recorded | 68fa753e1e209131cea66b68 | | Gemma-7B | 9.7 | Not recorded | 68fa753e1e209131cea66b68 | | EU | 12 | Not recorded | 68fa753e1e209131cea66b68 | | EU | 12.4 | Not recorded | 68fa753e1e209131cea66b68 | | RAG | 12.6 | Not recorded | 68fa753e1e209131cea66b68 | | LCM | 13.1 | Not recorded | 68fa753e1e209131cea66b68 | | RAG | 13.8 | Not recorded | 68fa753e1e209131cea66b68 | | LCM | 13.9 | Not recorded | 68fa753e1e209131cea66b68 | ### Gemma-7B ΔAPC (DeB): 1.1. Result ID: 699397123a6c31484917165d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO (APC-based DPO)":"False"}. This record does not assert independent reproduction. ### Experience Upload (EU) ΔAPC (DeB): 3.4. Result ID: 699397123a6c31484917165e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO (APC-based DPO)":"False"}. This record does not assert independent reproduction. ### Experience Upload (EU) ΔAPC (DeB): 3.8. Result ID: 699397123a6c314849171661. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO (APC-based DPO)":"True"}. This record does not assert independent reproduction. ### Retrieval-augmented Generation (RAG) ΔAPC (DeB): 4. Result ID: 699397123a6c314849171660. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO (APC-based DPO)":"False"}. This record does not assert independent reproduction. ### Long-context Memory (LCM) ΔAPC (DeB): 4.5. Result ID: 699397123a6c31484917165f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO (APC-based DPO)":"False"}. This record does not assert independent reproduction. ### Retrieval-augmented Generation (RAG) ΔAPC (DeB): 5.2. Result ID: 699397123a6c314849171663. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO (APC-based DPO)":"True"}. This record does not assert independent reproduction. ### Long-context Memory (LCM) ΔAPC (DeB): 5.3. Result ID: 699397123a6c314849171662. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO (APC-based DPO)":"True"}. This record does not assert independent reproduction. ### Gemma-7B ΔAPC (DeB): 9.7. Result ID: 699397123a6c314849171672. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO Status":"Without CPO"}. This record does not assert independent reproduction. ### EU ΔAPC (DeB): 12. Result ID: 699397123a6c314849171673. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO Status":"Without CPO"}. This record does not assert independent reproduction. ### EU ΔAPC (DeB): 12.4. Result ID: 699397123a6c314849171676. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO Status":"With CPO"}. This record does not assert independent reproduction. ### RAG ΔAPC (DeB): 12.6. Result ID: 699397123a6c314849171675. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO Status":"Without CPO"}. This record does not assert independent reproduction. ### LCM ΔAPC (DeB): 13.1. Result ID: 699397123a6c314849171674. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO Status":"Without CPO"}. This record does not assert independent reproduction. ### RAG ΔAPC (DeB): 13.8. Result ID: 699397123a6c314849171678. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO Status":"With CPO"}. This record does not assert independent reproduction. ### LCM ΔAPC (DeB): 13.9. Result ID: 699397123a6c314849171677. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"CPO Status":"With CPO"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.