# Text-to-Video Prompt Rewriting on VideoPhy 2 Source: [Text-to-Video Prompt Rewriting on VideoPhy 2]() Collection index: [llms.txt]() Data updated: 2026-03-05T07:32:43.826Z Evaluation of models on rewriting text prompts for video generation using the VideoPhy 2 dataset, assessed using SA and PC metrics. 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 | Text-to-Video Prompt Rewriting | | Dataset (including any stated split) | VideoPhy 2 | | Metric | SA Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 11 of 11 | | 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 - [SA Score](): Higher is better; unit: PERCENT. - [PC Score](): Higher is better; unit: PERCENT. - [SA & PC Score](): Higher is better; unit: PERCENT. - [Average SA Score](): Higher is better; unit: SCALAR. - [Average PC Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | SA Score | Result date | Report paper | | --- | --- | --- | --- | | DeepSeek-V3 | 48.4 | Not recorded | 69a90de72d603e6de3e4065a | | PhyPrompt-7B | 47.8 | Not recorded | 69a90de72d603e6de3e4065a | | GPT-4o | 47 | Not recorded | 69a90de72d603e6de3e4065a | | PhyPrompt-3B | 46.4 | Not recorded | 69a90de72d603e6de3e4065a | | PhyT2V | 45 | Not recorded | 69a90de72d603e6de3e4065a | | Qwen2.5-7B | 44.8 | Not recorded | 69a90de72d603e6de3e4065a | | Original Prompt | 43.4 | Not recorded | 69a90de72d603e6de3e4065a | | PhyPrompt-1.5B | 42.8 | Not recorded | 69a90de72d603e6de3e4065a | | Promptist | 41.2 | Not recorded | 69a90de72d603e6de3e4065a | | Qwen2.5-3B | 40 | Not recorded | 69a90de72d603e6de3e4065a | | Qwen2.5-1.5B | 37.6 | Not recorded | 69a90de72d603e6de3e4065a | ### DeepSeek-V3 SA Score: 48.4. Result ID: 69a9318d6dbb7075b9523ca1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B"}. This record does not assert independent reproduction. ### PhyPrompt-7B SA Score: 47.8. Result ID: 69a9318d6dbb7075b9523c9f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B","Model Size":"7B"}. This record does not assert independent reproduction. ### GPT-4o SA Score: 47. Result ID: 69a9318d6dbb7075b9523ca0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B"}. This record does not assert independent reproduction. ### PhyPrompt-3B SA Score: 46.4. Result ID: 69a9318d6dbb7075b9523c9d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B","Model Size":"3B"}. This record does not assert independent reproduction. ### PhyT2V SA Score: 45. Result ID: 69a9318d6dbb7075b9523c99. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B"}. This record does not assert independent reproduction. ### Qwen2.5-7B SA Score: 44.8. Result ID: 69a9318d6dbb7075b9523c9e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B","Model Size":"7B"}. This record does not assert independent reproduction. ### Original Prompt SA Score: 43.4. Result ID: 69a9318d6dbb7075b9523c97. [Reporting paper]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B"}. This record does not assert independent reproduction. ### PhyPrompt-1.5B SA Score: 42.8. Result ID: 69a9318d6dbb7075b9523c9b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B","Model Size":"1.5B"}. This record does not assert independent reproduction. ### Promptist SA Score: 41.2. Result ID: 69a9318d6dbb7075b9523c98. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B"}. This record does not assert independent reproduction. ### Qwen2.5-3B SA Score: 40. Result ID: 69a9318d6dbb7075b9523c9c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B","Model Size":"3B"}. This record does not assert independent reproduction. ### Qwen2.5-1.5B SA Score: 37.6. Result ID: 69a9318d6dbb7075b9523c9a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Generator":"CogVideoX-2B","Model Size":"1.5B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.