Research map / Label-efficient and robust learning / Recognition and long-tail
FedLAS: Feature-Modulated Bidirectional Label Smoothing for Neural Network Calibration
Thiru Thillai Nadarasar Bahavan, Sachith Seneviratne, Saman Halgamuge
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add FeDLaS adaptive smoothing plug-in | MbLS · Standard benchmarks (3 datasets, 2 ResNets) | mean ECE rank: 4.33 -> 1.83 | Improved |
| add FeDLaS plug-in | LS · Standard benchmarks | mean ECE rank: 7.00 -> 5.50 | Improved |
| add FeDLaS plug-in | LS · CIFAR-100 ResNet101 | ECE: 6.08 -> 7.48 | Hurt |
| add FeDLaS plug-in | LS · Fine-grained (mean) | mean ECE rank: 3.00 -> 1.50 | Improved |
| add FeDLaS plug-in | MbLS · Fine-grained ViT-B/16 @384 (Aircraft) | ECE: 21.89 -> 23.99 | Hurt |
| add FeDLaS plug-in | LS / MbLS · Standard benchmarks | mean top-1 accuracy: 79.69 -> 79.62; 79.66 -> 79.58 | No clear effect |
| add FeDLaS plug-in | LS · OOD detection CIFAR-10 vs SVHN | AUROC: 74.23 -> 85.04 | Improved |
| add FeDLaS plug-in | LS · OOD detection CIFAR-100 vs CIFAR-10 | AUROC: 72.33 -> 73.71, still below CE 74.82 | Mixed |
| add bidirectional gating (FeDLaS) | LS · CIFAR-10 decomposed ECE | U-ECE: 2.45 -> 0.40 | Improved |
| add FeDLaS plug-in | LS · CIFAR-100 decomposed ECE | O-ECE: 2.73 -> 3.70 | Hurt |
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.
- ACLS — ACLS: Adaptive and Conditional Label Smoothing for Network Calibration (ICCV 2023) · better on 3 of 3 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
- Towards Calibrated Multi-label Deep Neural Networks (CVPR 2024 · similarity 0.89)
- MaxSup: Overcoming Representation Collapse in Label Smoothing (NeurIPS 2025 · similarity 0.89)
- Training Deep Neural Networks with Virtual Smoothing Classes (AAAI 2025 · similarity 0.89)
- LaSCal: Label-Shift Calibration without target labels (NeurIPS 2024 · similarity 0.86)
- Dirichlet-Based Prediction Calibration for Learning with Noisy Labels (AAAI 2024 · similarity 0.86)
- NoiseGPT: Label Noise Detection and Rectification through Probability Curvature (NeurIPS 2024 · similarity 0.86)
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.
