Replace, Don't Expand: Mitigating Context Dilution in Multi-Hop RAG via Fixed-Budget Evidence Assembly

Documenti analoghi

• A Fixed-Budget, Cluster-Aware Standard for LLM-as-a-Judge Evaluation: A Multi-Hop RAG Stress Test
  di: Sartori, Camilo Chacón, et al.
  Pubblicazione: (2026)
• Don't Pay Attention
  di: Hammoud, Mohammad, et al.
  Pubblicazione: (2025)
• EfficientRAG: Efficient Retriever for Multi-Hop Question Answering
  di: Zhuang, Ziyuan, et al.
  Pubblicazione: (2024)
• Knowing You Don't Know: Learning When to Continue Search in Multi-round RAG through Self-Practicing
  di: Yang, Diji, et al.
  Pubblicazione: (2025)
• Honest AI: Fine-Tuning "Small" Language Models to Say "I Don't Know", and Reducing Hallucination in RAG
  di: Chen, Xinxi, et al.
  Pubblicazione: (2024)