# Contextual Knowledge Adaptation (Prior-Conflicting) on GENIUS Source: [Contextual Knowledge Adaptation (Prior-Conflicting) on GENIUS]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:59:45.776Z Evaluation of model performance on the Contextual Knowledge Adaptation task (specifically focusing on Prior-Conflicting scenarios) using the GENIUS dataset, measured by RC, VC, and AQ 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 | Contextual Knowledge Adaptation (Prior-Conflicting) | | Dataset (including any stated split) | GENIUS | | Metric | RC Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 13 of 13 | | 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 - [RC Score](): Higher is better; unit: PERCENT. - [VC Score](): Higher is better; unit: PERCENT. - [AQ Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | RC Score | Result date | Report paper | | --- | --- | --- | --- | | Nano Banana Pro | 52.97 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | GPT-Image | 43.5 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | SeeDream 4.5 | 40.1 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | Nano Banana | 35.5 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | SeeDream 4.0 | 30.69 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | Qwen-Image | 27.72 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | Emu3.5-Image | 26.24 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | FLUX.2-dev | 25.25 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | Omini-Gen2 | 23.76 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | Ours | 23.67 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | Bagel | 22.28 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | GLM-Image | 20.3 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | | NextStep-1 | 15.49 | Not recorded | 698d5ea8c97a0b3e3b5e3e7e | ### Nano Banana Pro RC Score: 52.97. Result ID: 6998c18c5698665ec424b47f. [Reporting paper]() Recorded evaluation configuration: {"Interleaved":true}. This record does not assert independent reproduction. ### GPT-Image RC Score: 43.5. Result ID: 6998c18c5698665ec424b481. [Reporting paper]() Recorded evaluation configuration: {"Interleaved":false}. This record does not assert independent reproduction. ### SeeDream 4.5 RC Score: 40.1. Result ID: 6998c18c5698665ec424b483. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Interleaved":false}. This record does not assert independent reproduction. ### Nano Banana RC Score: 35.5. Result ID: 6998c18c5698665ec424b480. [Reporting paper]() Recorded evaluation configuration: {"Interleaved":true}. This record does not assert independent reproduction. ### SeeDream 4.0 RC Score: 30.69. Result ID: 6998c18c5698665ec424b482. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Interleaved":false}. This record does not assert independent reproduction. ### Qwen-Image RC Score: 27.72. Result ID: 6998c18c5698665ec424b484. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Interleaved":false}. This record does not assert independent reproduction. ### Emu3.5-Image RC Score: 26.24. Result ID: 6998c18c5698665ec424b488. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Interleaved":false}. This record does not assert independent reproduction. ### FLUX.2-dev RC Score: 25.25. Result ID: 6998c18c5698665ec424b486. [Reporting paper]() Recorded evaluation configuration: {"Interleaved":false}. This record does not assert independent reproduction. ### Omini-Gen2 RC Score: 23.76. Result ID: 6998c18c5698665ec424b489. [Reporting paper]() Recorded evaluation configuration: {"Interleaved":false}. This record does not assert independent reproduction. ### Ours RC Score: 23.67. Result ID: 6998c18c5698665ec424b48b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Interleaved":true}. This record does not assert independent reproduction. ### Bagel RC Score: 22.28. Result ID: 6998c18c5698665ec424b48a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Interleaved":true}. This record does not assert independent reproduction. ### GLM-Image RC Score: 20.3. Result ID: 6998c18c5698665ec424b485. [Reporting paper]() Recorded evaluation configuration: {"Interleaved":false}. This record does not assert independent reproduction. ### NextStep-1 RC Score: 15.49. Result ID: 6998c18c5698665ec424b487. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Interleaved":false}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.