Research map / Label-efficient and robust learning / Recognition and long-tail
TransReID: Transformer-Based Object Re-Identification
Shuting He, Hao Luo, Pichao Wang, Fan Wang, Hao Li, Wei Jiang
This paper tried a common change: Add label smoothing
add label smoothing · MSMT17 · mAP: 61.0 -> 59.8 Hurt
What 190 accepted papers report for this change
- 127 improved
- 19 no clear effect
- 7 mixed
- 37 hurt
- Anti-Backdoor Coreset Selection via Cumulative Entropy ICML 2026 · add unlearning step with label smoothing · CIFAR10 WaNet · poisoning rate of coresetImproved
- When Distance Distracts: Representation Distance Bias in BT-Loss for Reward Models ICML 2026 · add label smoothing to BT loss · RewardBench Reasoning · accuracy: 75.41 -> 72.28Hurt
- Intrinsic Gradient Suppression for Label-Noise Prompt Tuning in Vision–Language Models ICML 2026 · apply label smoothing instead of plain CE · Average of 9 datasets, 40% sym noise · accuracy (%): 78.09 -> 78.39Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 jigsaw patch module (4 groups) (ablation) | ViT-B/16 baseline · MSMT17 | mAP: 61.0 -> 63.6 | Improved |
| add jigsaw patch module (4 groups) (ablation) | ViT-B/16 baseline · VeRi-776 | mAP: 78.2 -> 79.2 | Improved |
| add shift and shuffle rearrangement inside JPM (ablation) | JPM · MSMT17 | mAP: 63.1 -> 63.6 | Improved |
| add shift and shuffle rearrangement inside JPM (ablation) | JPM · VeRi-776 | mAP: 79.0 -> 79.2 | Improved |
| increase JPM groups from 1 to 4 (ablation) | JPM · MSMT17 | mAP: 62.9 -> 63.6 | Improved |
| add camera side information embedding (ablation) | ViT-B/16 baseline · MSMT17 | mAP: 61.0 -> 62.4 | Improved |
| encode camera and viewpoint jointly instead of summing embeddings (ablation) | SIE · VeRi-776 | mAP: 78.3 -> 79.6 | Improved |
| use overlapping patches with step size 12 (ablation) | ViT-B/16 baseline · MSMT17 | mAP: 61.0 -> 64.4 | Improved |
| combine JPM and SIE (ablation) | ViT-B/16 baseline · MSMT17 | mAP: 61.0 -> 64.9 | Improved |
| tune SIE weight lambda to 2.0 (ablation) | SIE · MSMT17 | mAP: 61.0 -> 63.0 | Improved |
| add position embeddings (ablation) | ViT-B/16 baseline · MSMT17 | mAP: 22.4 -> 61.0 | Improved |
| use Adam instead of SGD (ablation) | ViT-B/16 baseline · MSMT17 | mAP: 61.0 -> 37.4 | Hurt |
Closest papers by meaning
- TOP-ReID: Multi-Spectral Object Re-identification with Token Permutation (AAAI 2024 · similarity 0.88)
- Magic Tokens: Select Diverse Tokens for Multi-modal Object Re-Identification (CVPR 2024 · similarity 0.87)
- Alice Benchmarks: Connecting Real World Re-Identification with the Synthetic (ICLR 2024 · similarity 0.86)
- MDReID: Modality-Decoupled Learning for Any-to-Any Multi-Modal Object Re-Identification (NeurIPS 2025 · similarity 0.86)
- Catalyst for Clustering-Based Unsupervised Object Re-identification: Feature Calibration (AAAI 2024 · similarity 0.85)
- DeMo: Decoupled Feature-Based Mixture of Experts for Multi-Modal Object Re-Identification (AAAI 2025 · similarity 0.85)
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.
