mlsanityData to 7 Oct 2026

What really matters for person re-identification? A Mixture-of-Experts Framework for Semantic Attribute Importance

arXiv 2512.08697 ↗First posted 9 Dec 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

State-of-the-art person re-identification methods achieve impressive accuracy but remain largely opaque, leaving open the question: which high-level semantic attributes do these models actually rely on? We propose MoSAIC-ReID, a Mixture-of-Experts framework that systematically quantifies the importance of pedestrian attributes for re-identification. Our approach uses LoRA-based experts, each linked to a single attribute, and an oracle router that enables controlled attribution analysis. While MoSAIC-ReID achieves competitive performance on Market-1501 and DukeMTMC under the assumption that attribute annotations are available at test time, its primary value lies in providing a large-scale, quantitative study of attribute importance across intrinsic and extrinsic cues. Using generalized linear models, statistical tests, and feature-importance analyses, we reveal which attributes, such as clothing colors and intrinsic characteristics, contribute most strongly, while infrequent cues (e.g. accessories) have limited effect. This work offers a principled framework for interpretable ReID and highlights the requirements for integrating explicit semantic knowledge in practice. Code is available at https://github.com/psaltaath/MoSAIC-ReID

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 24 of its 41 references that are in the map. First posted on arXiv 9 Dec 2025.

Builds on

24 of 41 references.

References · 24

  1. 2017Attention Is All You Needcited by 59,932
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  4. 2021Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsitycited by 3,014
  5. 2024DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Modelscited by 834
  6. 2022Mixture-of-Experts with Expert Choice Routingcited by 611
  7. 2024DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understandingcited by 559
  8. 2018All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneouslycited by 364
  9. 2017Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitrocited by 280
  10. 2022CLIP-ReID: Exploiting Vision-Language Model for Image Re-Identification without Concrete Text Labelscited by 177
  11. 2024LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-trainingcited by 122
  12. 2024Higher Layers Need More LoRA Expertscited by 101
12 more
  1. 2017Improving Person Re-identification by Attribute and Identity Learningcited by 95
  2. 2023Beyond Appearance: a Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual Taskscited by 94
  3. 2021Generalizable Person Re-identification with Relevance-aware Mixture of Expertscited by 77
  4. 2024Flex-MoE: Modeling Arbitrary Modality Combination via the Flexible Mixture-of-Expertscited by 69
  5. 2021AAformer: Auto-Aligned Transformer for Person Re-Identificationcited by 48
  6. 2024DeMo: Decoupled Feature-Based Mixture of Experts for Multi-Modal Object Re-Identificationcited by 29
  7. 2025IDEA: Inverted Text with Cooperative Deformable Aggregation for Multi-modal Object Re-Identificationcited by 27
  8. 2025SPMTrack: Spatio-Temporal Parameter-Efficient Fine-Tuning with Mixture of Experts for Scalable Visual Trackingcited by 23
  9. 2023Multi-Prompts Learning with Cross-Modal Alignment for Attribute-based Person Re-Identificationcited by 20
  10. 2024Lifelong Knowledge Editing for Vision Language Models with Low-Rank Mixture-of-Expertscited by 14
  11. 2025MoEE: Mixture of Emotion Experts for Audio-Driven Portrait Animationcited by 12
  12. 2024A Culturally-Aware Benchmark for Person Re-Identification in Modest Attirecited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Attribute Guidance With Inherent Pseudo-label For Occluded Person Re-identification
  2. 2025FlexiReID: Adaptive Mixture of Expert for Multi-Modal Person Re-Identification
  3. 2025Find Them All: Unveiling MLLMs for Versatile Person Re-identification
  4. 2024When Large Vision-Language Models Meet Person Re-Identification
  5. 2025ChatReID: Open-ended Interactive Person Retrieval via Hierarchical Progressive Tuning for Vision Language Models
  6. 2025What Makes You Unique? Attribute Prompt Composition for Object Re-Identification
  7. 2026Vision-Language Attribute Disentanglement and Reinforcement for Lifelong Person Re-Identification
  8. 2025Modeling Thousands of Human Annotators for Generalizable Text-to-Image Person Re-identification