# Instruction Tuning on Vicuna Source: [Instruction Tuning on Vicuna]() Collection index: [llms.txt]() Data updated: 2026-05-12T08:59:40.672Z Evaluation of model performance on Instruction Tuning using the Vicuna dataset, measured by RougeL score. 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 | Instruction Tuning | | Dataset (including any stated split) | Vicuna | | Metric | RougeL Score | | 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 - [RougeL Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | RougeL Score | Result date | Report paper | | --- | --- | --- | --- | | ARMADA | 18.73 | Not recorded | 69b2488ed030084c22e2f8d4 | | ARMADA | 18.67 | Not recorded | 69b2488ed030084c22e2f8d4 | | ARMADA | 18.44 | Not recorded | 69b2488ed030084c22e2f8d4 | | ARMADA | 18.16 | Not recorded | 69b2488ed030084c22e2f8d4 | | ARMADA | 18.15 | Not recorded | 69b2488ed030084c22e2f8d4 | | ARMADA | 18.15 | Not recorded | 69b2488ed030084c22e2f8d4 | | ARMADA | 18.01 | Not recorded | 69b2488ed030084c22e2f8d4 | | ARMADA | 17.92 | Not recorded | 69b2488ed030084c22e2f8d4 | | ARMADA | 17.84 | Not recorded | 69b2488ed030084c22e2f8d4 | | SeqKD | 17.74 | Not recorded | 69b2488ed030084c22e2f8d4 | | Undistilled | 17.59 | Not recorded | 69b2488ed030084c22e2f8d4 | | S2FT | 7.94 | Not recorded | 6a02b430dbc836b87d8247ce | | FourierFT | 7.92 | Not recorded | 6a02b430dbc836b87d8247ce | | LoRA | 7.89 | Not recorded | 6a02b430dbc836b87d8247ce | | LoRA | 7.89 | Not recorded | 6a02b430dbc836b87d8247ce | | S2FT | 7.5 | Not recorded | 6a02b430dbc836b87d8247ce | | FourierFT | 7.49 | Not recorded | 6a02b430dbc836b87d8247ce | | LoRA | 7.38 | Not recorded | 6a02b430dbc836b87d8247ce | | LoRA | 7.35 | Not recorded | 6a02b430dbc836b87d8247ce | ### ARMADA RougeL Score: 18.73. Result ID: 69b285886dbb7075b95d6e89. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Teacher":"text-to-image","Loss Function":"L\_cosine","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### ARMADA RougeL Score: 18.67. Result ID: 69b285886dbb7075b95d6e91. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Teacher":"text-to-audio","Loss Function":"L\_elementwise","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### ARMADA RougeL Score: 18.44. Result ID: 69b285886dbb7075b95d6e8c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Teacher":"text-to-video","Loss Function":"L\_cosine","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### ARMADA RougeL Score: 18.16. Result ID: 69b285886dbb7075b95d6e8b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Teacher":"text-to-image","Loss Function":"L\_elementwise","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### ARMADA RougeL Score: 18.15. Result ID: 69b285886dbb7075b95d6e8a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Teacher":"text-to-image","Loss Function":"L\_euclid","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### ARMADA RougeL Score: 18.15. Result ID: 69b285886dbb7075b95d6e8f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Teacher":"text-to-audio","Loss Function":"L\_cosine","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### ARMADA RougeL Score: 18.01. Result ID: 69b285886dbb7075b95d6e90. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Teacher":"text-to-audio","Loss Function":"L\_euclid","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### ARMADA RougeL Score: 17.92. Result ID: 69b285886dbb7075b95d6e8d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Teacher":"text-to-video","Loss Function":"L\_euclid","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### ARMADA RougeL Score: 17.84. Result ID: 69b285886dbb7075b95d6e8e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Teacher":"text-to-video","Loss Function":"L\_elementwise","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### SeqKD RougeL Score: 17.74. Result ID: 69b285886dbb7075b95d6e88. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Teacher":"language-only","Distillation Type":"SeqKD","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### Undistilled RougeL Score: 17.59. Result ID: 69b285886dbb7075b95d6e87. [Reporting paper]() Recorded evaluation configuration: {"Teacher":"-","Distillation Type":"undistilled","Student Model":"LLaMA-3.2-3B"}. This record does not assert independent reproduction. ### S2FT RougeL Score: 7.94. Result ID: 6a02ebe447f34ce924e5d5b9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA2-13B","# Trainable Parameters":"0.08M"}. This record does not assert independent reproduction. ### FourierFT RougeL Score: 7.92. Result ID: 6a02ebe447f34ce924e5d5b8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA2-13B","# Trainable Parameters":"0.08M"}. This record does not assert independent reproduction. ### LoRA RougeL Score: 7.89. Result ID: 6a02ebe447f34ce924e5d5b6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA2-13B","# Trainable Parameters":"250.3M","Updating layers other than lm\_head":"true"}. This record does not assert independent reproduction. ### LoRA RougeL Score: 7.89. Result ID: 6a02ebe447f34ce924e5d5b7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA2-13B","# Trainable Parameters":"52.4M","Updating layers other than lm\_head":"false"}. This record does not assert independent reproduction. ### S2FT RougeL Score: 7.5. Result ID: 6a02ebe447f34ce924e5d5b5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA2-7B","# Trainable Parameters":"0.064M"}. This record does not assert independent reproduction. ### FourierFT RougeL Score: 7.49. Result ID: 6a02ebe447f34ce924e5d5b4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA2-7B","# Trainable Parameters":"0.064M"}. This record does not assert independent reproduction. ### LoRA RougeL Score: 7.38. Result ID: 6a02ebe447f34ce924e5d5b2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA2-7B","# Trainable Parameters":"159.9M","Updating layers other than lm\_head":"true"}. This record does not assert independent reproduction. ### LoRA RougeL Score: 7.35. Result ID: 6a02ebe447f34ce924e5d5b3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA2-7B","# Trainable Parameters":"33.5M","Updating layers other than lm\_head":"false"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.