Text or Pixels? It Takes Half: On the Token Efficiency of Visual Text Inputs in Multimodal LLMs

Documenti analoghi

• Beyond Text: Optimizing RAG with Multimodal Inputs for Industrial Applications
  di: Riedler, Monica, et al.
  Pubblicazione: (2024)
• CSRP: Chain-of-Thought Reasoning for Chinese Text Correction via Reinforcement Learning with Efficiency-Aware Rewards
  di: Tian, Wei, et al.
  Pubblicazione: (2026)
• Chunk-Distilled Language Modeling
  di: Li, Yanhong, et al.
  Pubblicazione: (2024)
• Information Capacity: Evaluating the Efficiency of Large Language Models via Text Compression
  di: Yuan, Cheng, et al.
  Pubblicazione: (2025)
• Unsupervised Text Style Transfer via LLMs and Attention Masking with Multi-way Interactions
  di: Pan, Lei, et al.
  Pubblicazione: (2024)