Studying Image Tokenizers as Visual Languages in Unified Multimodal Models

  • 类型：arxiv
  • 标识：2609.09143
  • 链接：https://arxiv.org/abs/2609.09143
  • 主分类：multimodal
  • 形态：method
  • 被引：0
  • 被引来源：Semantic Scholar + OpenAlex
  • S2被引：0
  • OpenAlex被引：0
  • 影响力被引：0
  • TLDR：A controlled pure-autoregressive testbed is built and track task-specific validation losses during multimodal continual pretraining across text, image, text-to-image (T2I), and image-to-text (I2T) prediction, showing that better reconstruction does not necessarily yield lower task-specific losses or stronger downstream performance, and that image tokenizer choice can affect text modeling under joint optimization.
  • OpenAlex ID：W7212079998
  • OpenAlex DOI：10.48550/arxiv.2609.09143
  • DOI：10.48550/arxiv.2609.09143
  • DOI来源：OpenAlex
  • 开放获取：green
  • 开放获取链接：https://arxiv.org/pdf/2609.09143
  • OpenAlex更新：2026-09-13
  • 副分类：evaluation
  • 待LLM分类：否
  • 标题中文：在统一多模态模型中将图像 tokenizer 视为视觉语言的研究
  • TLDR中文：构建一个受控的纯自回归测试床，在文本、图像、文生图（T2I）和图生文（I2T）预测的多模态持续预训练中跟踪任务特定验证损失，表明更好的重建并不一定带来更低的任务特定损失或更强的下游性能，且图像分词器的选择在联合优化下会影响文本建模。
  • 来源文件：
  • /inbox/tom/_candidates/2026-09-12-agent-rag-longcontext-candidates.json
  • [OpenAlex backfill]
  • /inbox/tom/_candidates/2026-09-13-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-09-14-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-09-14-agent-memory-tool-use-candidates.json
  • [S2 enrich]