When Hard Negatives Hurt: Bridging the Generative-Discriminative Gap in Hard Negative Synthesis for Retrieval

Documenti analoghi

• Hard Negative Mining for Domain-Specific Retrieval in Enterprise Systems
  di: Meghwani, Hansa, et al.
  Pubblicazione: (2025)
• Length-Adaptive Interest Network for Balancing Long and Short Sequence Modeling in CTR Prediction
  di: Zhang, Zhicheng, et al.
  Pubblicazione: (2026)
• SyNeg: LLM-Driven Synthetic Hard-Negatives for Dense Retrieval
  di: Li, Xiaopeng, et al.
  Pubblicazione: (2024)
• MeVer at CheckThat! 2026: Cluster-Aware Hard-Negative Mining for Multilingual Scientific-Source Retrieval
  di: Bakagianni, Juli, et al.
  Pubblicazione: (2026)
• ESANS: Effective and Semantic-Aware Negative Sampling for Large-Scale Retrieval Systems
  di: Xing, Haibo, et al.
  Pubblicazione: (2025)