AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Quantifying and Optimizing Simplicity via Polynomial Representations

Tianren Zhang, Xiangxin Li, Minghao Xiao, Guanyu Chen, Feng Chen

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add effective-degree (ED) regularization with label anchoringViT-Tiny · CIFAR-10 (from scratch)top-1 accuracy (%): 87.80 -> 90.82Improved
add SAM, ASAM or Jacobian regularizationViT-Tiny · CIFAR-10 (from scratch)top-1 accuracy (%): 87.80 -> 87.85 (SAM)No clear effect
remove label anchoring from ED (ablation)ED regularization · CIFAR-10 (ViT-Tiny)top-1 accuracy (%): 90.82 -> 90.00Hurt
add ED regularization during CLIP fine-tuningCLIP ViT-B/32 · ImageNet ID and average OODaccuracy (%): ID 76.20 -> 77.14; avg OOD 44.04 -> 45.31Improved
add ED regularization during CLIP fine-tuningCLIP ViT-B/16 · ImageNet ID and average OODaccuracy (%): ID 81.35 -> 82.19; avg OOD 53.69 -> 55.29Improved
add ED regularization to ViT-S/16 trained from scratch on ImageNetViT-S/16 · ImageNet (original recipe)top-1 accuracy (%): 71.37 -> 72.76Improved
add ED regularization to ViT-S/16 with strong recipeViT-S/16 · ImageNet (strong recipe)top-1 accuracy (%): 74.42 -> 75.01Improved
add ED regularization with embedding-space pathsBERT-base · GLUE RTE, MRPC, CoLAaccuracy / Matthews corr.: RTE 70.28 -> 71.12; MRPC 86.74 -> 87.66; CoLA 62.31 -> 62.45Improved
add embedding mixupBERT-base · GLUE RTE, MRPC, CoLAaccuracy / Matthews corr.: MRPC 86.74 -> 86.30; CoLA 62.31 -> 61.04; RTE unchanged 70.28Hurt
add ED regularization to PPO actorPPO · Procgen Dodgeball, Fruitbot, Jumper, StarPilot (unseen levels)test scoreImproved
replace real-image interpolation paths with random noise (ablation)ED · CIFAR-10 (Appendix C.1)correlation and regularization performanceHurt
replace Chebyshev basis with Legendre basis (ablation)ED · CIFAR-10 (Appendix C.2)performanceNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy