Co-occurring keywords · papers mentioning both
Papers
Counteracting the Matthew Effect in Self-Improvement of LVLMs through Head-Tail Re-balancing
ACL 2026
Large Language Models Reflect Human Citation Patterns with a Heightened Citation Bias
NAACL 2025
HyCoRec: Hypergraph-Enhanced Multi-Preference Learning for Alleviating Matthew Effect in Conversational Recommendation
ACL 2024
Mitigating Matthew Effect: Multi-Hypergraph Boosted Multi-Interest Self-Supervised Learning for Conversational Recommendation
EMNLP 2024