SeongKu Kang

CV of Failure

This is a record of the failures I have experienced along the way.

I am deeply grateful to all my colleagues who have gone through these failures with me and shared the experience.

PAPER REJECTIONS

  1. Concepts Complement Dense Semantics: Learning Compact Sparse Spaces for Text-Image Retrieval

    • rejected from CIKM 2026.
    • accepted to KEIR@CIKM 2026.
  2. PEARL: Front-Loading Relational Chains for Multi-Hop Table Retrieval

    • rejected from SIGIR 2026.
    • accepted to EMNLP 2026.
  3. Beyond Observed Auxiliary Relations: Environment-Conditioned Modeling for Multi-Behavior Recommendation

    • rejected from KDD 2026.
    • accepted to CIKM 2026.
  4. Dynamic Spectral Decoupling for Time-Series Forecasting

    • rejected from NeurIPS 2025.
    • rejected from ICLR 2026.
    • rejected from KDD 2026.
    • accepted to CIKM 2026.
  5. TRACER: Balancing Stability-Plasticity-Cognitivity Trilemma for LLM Enhanced Continual Recommendation

    • rejected from KDD 2026.
    • accepted to CIKM 2026.
  6. CoRank: LLM-Based Compact Reranking with Document Features for Scientific Retrieval

    • rejected from COLM 2025.
    • accepted to KDD 2026.
  7. Why These Documents? Explainable Generative Retrieval with Hierarchical Category Paths

    • rejected from SIGIR 2025.
    • rejected from WWW 2025.
    • rejected from AAAI 2026.
    • accepted to ACL 2026.
  8. FLAME: Condensing Ensemble Diversity into a Single Network for Efficient Sequential Recommendation

    • rejected from CIKM 2025.
    • rejected from WWW 2026.
    • accepted to SIGIR 2026.
  9. MVIGER: Multi-View Variational Integration of Complementary Knowledge for Generative Recommender

    • rejected from CIKM 2024.
    • rejected from AAAI 2025.
    • rejected from SIGIR 2025.
    • rejected from WWW 2026.
    • accepted to SIGIR 2026.
  10. Filling the Gaps: Selective Knowledge Augmentation for LLM Recommenders

    • rejected from KDD 2026.
    • accepted to SIGIR 2026.
  11. SPRINT: Scalable and Predictive Intent Refinement for LLM-Enhanced Session-based Recommendation

    • rejected from KDD 2026.
    • accepted to SIGIR 2026.
  12. BPL: Bias-adaptive Preference Distillation Learning for Recommender System

    • rejected from NeurIPS 2023.
    • rejected from WWW 2024.
    • rejected from SIGIR 2024.
    • rejected from CIKM 2024.
    • accepted to TKDE 2026.
  13. Towards Unified and Adaptive Cross-Domain Collaborative Filtering via Graph Signal Processing

    • rejected from CIKM 2024.
    • rejected from KDD 2025.
    • rejected from SIGIR 2025.
    • rejected from CIKM 2025.
    • rejected from AAAI 2026.
    • Archived.
  14. Leveraging Historical and Current Interests for Continual Sequential Recommendation

    • rejected from SIGIR 2025.
    • rejected from WWW 2025.
    • rejected from CIKM 2025.
    • accepted to WSDM 2026.
  15. Chain-of-Factors Paper-Reviewer Matching

    • rejected from KDD 2025.
    • accepted to WWW 2025.
  16. Uncertainty Quantification and Decomposition for LLM-based Recommendation

    • rejected from CIKM 2024.
    • rejected from KDD 2025.
    • accepted to WWW 2025.
  17. Personalized Preference Reasoning with Large Language Models for Accurate and Explainable Recommendation

    • rejected from WWW 2024.
    • rejected from KDD 2025.
    • accepted to SIGIR 2025.
  18. Unsupervised Robust Cross-Lingual Entity Alignment via Neighbor Triple Matching with Entity and Relation Texts

    • rejected from WWW 2024.
    • rejected from KDD 2024.
    • rejected from CIKM 2024.
    • accepted to WSDM 2025.
  19. Continual Collaborative Distillation for Recommender System

    • rejected from CIKM 2024.
    • accepted to KDD 2024.
  20. Improving Matrix Completion by Exploiting Rating Ordinality in Graph Neural Networks

    • rejected from CIKM 2022.
    • rejected from CIKM 2023.
    • rejected from WWW 2024.
    • Archived.
    • We gave up. ๐Ÿ˜ข
  21. Top-Personalized-K Recommendation

    • rejected from KDD 2023.
    • rejected from CIKM 2023.
    • accepted to WWW 2024.
  22. Multi-Domain Recommendation to Attract Users via Domain Preference Modeling

    • rejected from CIKM 2023.
    • accepted to AAAI 2024.
  23. Mitigating Viewpoint Sensitivity of Self-Supervised One-Class Classifiers

    • rejected from ICLR 2021.
    • rejected from ECML/PKDD 2021.
    • accepted to Information Sciences 2022.
  24. On Cross-feedback Transferability of Explicit Ranking Information for Implicit Collaborative Filtering

    • rejected from KDD 2022.
    • We gave up. ๐Ÿ˜ข
  25. Unsupervised Proxy Selection for Session-based Recommender Systems

    • rejected from WWW 2021.
    • accepted to SIGIR 2021.
  26. Bootstrapping User and Item Representations for One-Class Collaborative Filtering

    • rejected from WWW 2021.
    • accepted to SIGIR 2021.
  27. Learning Heterogeneous Temporal Patterns of User Preference for Timely Recommendation

    • rejected from KDD 2020.
    • rejected from CIKM 2020.
    • accepted to WWW 2021.
  28. Deep Semi-Supervised Anomaly Detection with Laplacian Smoothing

    • rejected from KDD 2020.
    • We gave up. ๐Ÿ˜ข

GRANT, FELLOWSHIP, AND RESEARCH PROJECT REJECTIONS

  • KISTI ์Šˆํผ์ปด 6ํ˜ธ๊ธฐ R&Dํ˜์‹  ํ”„๋กœ๊ทธ๋žจ (1์ฐจ) ํƒˆ๋ฝ (2026)
  • ํฌ์Šคํ…โ€“๊ณ ๋ ค๋Œ€ ๊ณต๋™์—ฐ๊ตฌ๊ณผ์ œ Co-Seed Funding ํƒˆ๋ฝ (2026)
  • ํ•œ๊ตญ์ฝ˜ํ…์ธ ์ง„ํฅ์› ๋ฌธํ™”์˜ˆ์ˆ ์˜จํ†จ๋กœ์ง€๊ธฐ๋ฐ˜LLM์—ฐ๊ณ„๊ธฐ์ˆ ๊ฐœ๋ฐœ ํƒˆ๋ฝ (2026)
  • ๊ณ ๋ ค๋Œ€ํ•™๊ตโ€“DGIST ๊ธฐํš ์—ฐ๊ตฌ๊ณผ์ œ(Co-Seed Funding) ํƒˆ๋ฝ (2026)
  • ํ•œ๊ตญ์—ฐ๊ตฌ์žฌ๋‹จ AI ๊ธฐ๋ฐ˜ ๋Œ€ํ•™ ๊ณผํ•™๊ธฐ์ˆ  ํ˜์‹ ์‚ฌ์—…(AI4S&T) ํƒˆ๋ฝ (2026)
  • NIPA/KAIT ๊ณ ์„ฑ๋Šฅ์ปดํ“จํŒ… GPU ์ง€์›์‚ฌ์—… ํƒˆ๋ฝ (2025)
  • ํ•œ๊ตญ์—ฐ๊ตฌ์žฌ๋‹จ ์‹ ์ง„์—ฐ๊ตฌ ํƒˆ๋ฝ (2025)
  • Meta internship ํƒˆ๋ฝ (2021)
  • Microsoft Research Asia Fellowship ํƒˆ๋ฝ (2021)
  • NAVER PhD Fellowship ํƒˆ๋ฝ (2020)