# Image Classification on ILSVRC 2015 (val) Source: [Image Classification on ILSVRC 2015 (val)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:06:53.153Z Evaluation of Image Classification performance on the validation split of the ILSVRC 2015 dataset, measured using Top-1 and Top-5 accuracy metrics across different class subsets (C\_large, C\_few). 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 | Image Classification | | Dataset (including any stated split) | ILSVRC 2015 (val) | | Metric | Top-1 Acc (C\_large) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 33 of 33 | | 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 - [Top-1 Acc (C\_large)](): Higher is better; unit: PERCENT. - [Top-5 Acc (C\_large)](): Higher is better; unit: PERCENT. - [Top-1 Acc (C\_few)](): Higher is better; unit: PERCENT. - [Top-5 Acc (C\_few)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Top-1 Acc (C\_large) | Result date | Report paper | | --- | --- | --- | --- | | Learning like a Child | 76.71 | Not recorded | 68fa63d01e209131cea00e1c | | Learning like a Child | 76.68 | Not recorded | 68fa63d01e209131cea00e1c | | Learning like a Child | 76.55 | Not recorded | 68fa63d01e209131cea00e1c | | Fine-Tuned ResNet | 76.28 | Not recorded | 68fa63d01e209131cea00e1c | | Fine-Tuned ResNet | 76.27 | Not recorded | 68fa63d01e209131cea00e1c | | Fine-Tuned ResNet | 76.25 | Not recorded | 68fa63d01e209131cea00e1c | | Ours-phi2\* | 75.63 | Not recorded | 68fa63d01e209131cea00e1c | | Ours-phi2\* | 75.44 | Not recorded | 68fa63d01e209131cea00e1c | | Ours-phi2\* | 75.34 | Not recorded | 68fa63d01e209131cea00e1c | | Ours-phi2 | 74.17 | Not recorded | 68fa63d01e209131cea00e1c | | Ours-phi2 | 73.43 | Not recorded | 68fa63d01e209131cea00e1c | | Ours-phi2 | 72.98 | Not recorded | 68fa63d01e209131cea00e1c | | Ours-phi1 | 72.56 | Not recorded | 68fa63d01e209131cea00e1c | | Ours-phi1 | 71.94 | Not recorded | 68fa63d01e209131cea00e1c | | Ours-phi1 | 71.56 | Not recorded | 68fa63d01e209131cea00e1c | | NN + Cosine | 71.54 | Not recorded | 68fa63d01e209131cea00e1c | | NN + Cosine | 71.54 | Not recorded | 68fa63d01e209131cea00e1c | | NN + Cosine | 71.54 | Not recorded | 68fa63d01e209131cea00e1c | | Triplet Network | 70.47 | Not recorded | 68fa63d01e209131cea00e1c | | Triplet Network | 70.47 | Not recorded | 68fa63d01e209131cea00e1c | | Triplet Network | 70.47 | Not recorded | 68fa63d01e209131cea00e1c | | NN + Cosine | 67.68 | Not recorded | 68fa63d01e209131cea00e1c | | NN + Cosine | 67.66 | Not recorded | 68fa63d01e209131cea00e1c | | NN + Cosine | 67.65 | Not recorded | 68fa63d01e209131cea00e1c | | Triplet Network | 66.64 | Not recorded | 68fa63d01e209131cea00e1c | | Triplet Network | 66.63 | Not recorded | 68fa63d01e209131cea00e1c | | Triplet Network | 66.62 | Not recorded | 68fa63d01e209131cea00e1c | | NN + Cosine | 61.11 | Not recorded | 68fa63d01e209131cea00e1c | | NN + Cosine | 61.04 | Not recorded | 68fa63d01e209131cea00e1c | | NN + Cosine | 60.97 | Not recorded | 68fa63d01e209131cea00e1c | | Triplet Network | 60.09 | Not recorded | 68fa63d01e209131cea00e1c | | Triplet Network | 60.02 | Not recorded | 68fa63d01e209131cea00e1c | | Triplet Network | 59.97 | Not recorded | 68fa63d01e209131cea00e1c | ### Learning like a Child Top-1 Acc (C\_large): 76.71. Result ID: 698c03a8f729bc7f68e7d215. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":1,"Fine-tuning":true}. This record does not assert independent reproduction. ### Learning like a Child Top-1 Acc (C\_large): 76.68. Result ID: 698c03a8f729bc7f68e7d220. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":2,"Fine-tuning":true}. This record does not assert independent reproduction. ### Learning like a Child Top-1 Acc (C\_large): 76.55. Result ID: 698c03a8f729bc7f68e7d22b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":3,"Fine-tuning":true}. This record does not assert independent reproduction. ### Fine-Tuned ResNet Top-1 Acc (C\_large): 76.28. Result ID: 698c03a8f729bc7f68e7d214. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":1,"Fine-tuning":true}. This record does not assert independent reproduction. ### Fine-Tuned ResNet Top-1 Acc (C\_large): 76.27. Result ID: 698c03a8f729bc7f68e7d21f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":2,"Fine-tuning":true}. This record does not assert independent reproduction. ### Fine-Tuned ResNet Top-1 Acc (C\_large): 76.25. Result ID: 698c03a8f729bc7f68e7d22a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":3,"Fine-tuning":true}. This record does not assert independent reproduction. ### Ours-phi2\* Top-1 Acc (C\_large): 75.63. Result ID: 698c03a8f729bc7f68e7d218. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":1,"Fine-tuning":false}. This record does not assert independent reproduction. ### Ours-phi2\* Top-1 Acc (C\_large): 75.44. Result ID: 698c03a8f729bc7f68e7d223. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":2,"Fine-tuning":false}. This record does not assert independent reproduction. ### Ours-phi2\* Top-1 Acc (C\_large): 75.34. Result ID: 698c03a8f729bc7f68e7d22e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":3,"Fine-tuning":false}. This record does not assert independent reproduction. ### Ours-phi2 Top-1 Acc (C\_large): 74.17. Result ID: 698c03a8f729bc7f68e7d217. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":1,"Fine-tuning":false}. This record does not assert independent reproduction. ### Ours-phi2 Top-1 Acc (C\_large): 73.43. Result ID: 698c03a8f729bc7f68e7d222. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":2,"Fine-tuning":false}. This record does not assert independent reproduction. ### Ours-phi2 Top-1 Acc (C\_large): 72.98. Result ID: 698c03a8f729bc7f68e7d22d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":3,"Fine-tuning":false}. This record does not assert independent reproduction. ### Ours-phi1 Top-1 Acc (C\_large): 72.56. Result ID: 698c03a8f729bc7f68e7d216. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":1,"Fine-tuning":false}. This record does not assert independent reproduction. ### Ours-phi1 Top-1 Acc (C\_large): 71.94. Result ID: 698c03a8f729bc7f68e7d221. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":2,"Fine-tuning":false}. This record does not assert independent reproduction. ### Ours-phi1 Top-1 Acc (C\_large): 71.56. Result ID: 698c03a8f729bc7f68e7d22c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":3,"Fine-tuning":false}. This record does not assert independent reproduction. ### NN + Cosine Top-1 Acc (C\_large): 71.54. Result ID: 698c03a8f729bc7f68e7d20e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":1,"Fine-tuning":false}. This record does not assert independent reproduction. ### NN + Cosine Top-1 Acc (C\_large): 71.54. Result ID: 698c03a8f729bc7f68e7d219. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":2,"Fine-tuning":false}. This record does not assert independent reproduction. ### NN + Cosine Top-1 Acc (C\_large): 71.54. Result ID: 698c03a8f729bc7f68e7d224. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":3,"Fine-tuning":false}. This record does not assert independent reproduction. ### Triplet Network Top-1 Acc (C\_large): 70.47. Result ID: 698c03a8f729bc7f68e7d211. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":1,"Fine-tuning":false}. This record does not assert independent reproduction. ### Triplet Network Top-1 Acc (C\_large): 70.47. Result ID: 698c03a8f729bc7f68e7d21c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":2,"Fine-tuning":false}. This record does not assert independent reproduction. ### Triplet Network Top-1 Acc (C\_large): 70.47. Result ID: 698c03a8f729bc7f68e7d227. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"100%","samples per category (D\_few)":3,"Fine-tuning":false}. This record does not assert independent reproduction. ### NN + Cosine Top-1 Acc (C\_large): 67.68. Result ID: 698c03a8f729bc7f68e7d20f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"10%","samples per category (D\_few)":1,"Fine-tuning":false}. This record does not assert independent reproduction. ### NN + Cosine Top-1 Acc (C\_large): 67.66. Result ID: 698c03a8f729bc7f68e7d21a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"10%","samples per category (D\_few)":2,"Fine-tuning":false}. This record does not assert independent reproduction. ### NN + Cosine Top-1 Acc (C\_large): 67.65. Result ID: 698c03a8f729bc7f68e7d225. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"10%","samples per category (D\_few)":3,"Fine-tuning":false}. This record does not assert independent reproduction. ### Triplet Network Top-1 Acc (C\_large): 66.64. Result ID: 698c03a8f729bc7f68e7d212. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"10%","samples per category (D\_few)":1,"Fine-tuning":false}. This record does not assert independent reproduction. ### Triplet Network Top-1 Acc (C\_large): 66.63. Result ID: 698c03a8f729bc7f68e7d21d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"10%","samples per category (D\_few)":2,"Fine-tuning":false}. This record does not assert independent reproduction. ### Triplet Network Top-1 Acc (C\_large): 66.62. Result ID: 698c03a8f729bc7f68e7d228. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"10%","samples per category (D\_few)":3,"Fine-tuning":false}. This record does not assert independent reproduction. ### NN + Cosine Top-1 Acc (C\_large): 61.11. Result ID: 698c03a8f729bc7f68e7d210. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"1%","samples per category (D\_few)":1,"Fine-tuning":false}. This record does not assert independent reproduction. ### NN + Cosine Top-1 Acc (C\_large): 61.04. Result ID: 698c03a8f729bc7f68e7d21b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"1%","samples per category (D\_few)":2,"Fine-tuning":false}. This record does not assert independent reproduction. ### NN + Cosine Top-1 Acc (C\_large): 60.97. Result ID: 698c03a8f729bc7f68e7d226. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"1%","samples per category (D\_few)":3,"Fine-tuning":false}. This record does not assert independent reproduction. ### Triplet Network Top-1 Acc (C\_large): 60.09. Result ID: 698c03a8f729bc7f68e7d213. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"1%","samples per category (D\_few)":1,"Fine-tuning":false}. This record does not assert independent reproduction. ### Triplet Network Top-1 Acc (C\_large): 60.02. Result ID: 698c03a8f729bc7f68e7d21e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"1%","samples per category (D\_few)":2,"Fine-tuning":false}. This record does not assert independent reproduction. ### Triplet Network Top-1 Acc (C\_large): 59.97. Result ID: 698c03a8f729bc7f68e7d229. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"D\_large percentage":"1%","samples per category (D\_few)":3,"Fine-tuning":false}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.