Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 30, 2026International Journal of Coal Preparation and Utilization

MCL: Mamba-Based Contrastive Learning method for coal gangue identification

View Full Paper
Ask AI
Bookmark
Share

Authors

HQHaifeng QiuChina University of Mining and TechnologyZZZipeng ZhangChina University of Mining and TechnologyBZBo ZhangSouth China National Centre of Metrology

Discussion

Loading...

Member takes

Implication

Randomized trial demonstrates effective coal-gangue identification using a novel contrastive learning framework.

Key Points

  • The aim is to improve the identification of coal versus gangue in mining operations using a new learning framework.
  • Developed a contrastive learning framework with dual attention and Mamba modules for feature extraction.
  • Used self-attention branches to enhance feature representation and model class-specific similarities.
  • Conducted ablation and comparative experiments on a simulation dataset to assess effectiveness.
  • Achieved 95.03% accuracy and 94.95% F1-score, outperforming all baseline methods.
  • Obtained 94.78% precision and 95.16% recall, indicating strong model performance.
  • Kappa and MCC scores reached 89.89% and 89.94%, highlighting robust generalization capabilities.

Cite This Study

Qiu et al. (2026) studied this question.

synapsesocial.com/papers/6a6af51160e2b924d3ea0b17https://doi.org/10.1080/19392699.2026.2708066
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1WFTNet: Exploiting Global and Local Periodicity in Long-Term Time Series Forecasting2024 · 38 citations
  2. 2Positive-Augmented Contrastive Learning for Image and Video Captioning Evaluation2023 · 43 citations
  3. 3Divergence measures based on the Shannon entropy1991 · 5,291 citations
  4. 4Research on the fine identification of coal types and gangue based on X-ray diffraction principle2022 · 11 citations
  5. 5Enhancing spatiotemporal prediction through the integration of Mamba state space models and Diffusion Transformers2025 · 40 citations