T
traeai
Sign in

概念

OCR

别名:光学字符识别

Optical Character Recognition,光学字符识别

已跟踪 3 条高相关材料

TraeAI 观察

相关材料

已收录 3 条与 OCR 相关的内容,按评分排序。

Direct Preference Optimization Beyond Chatbots

Direct Preference Optimization Beyond Chatbots

Hugging Face Blog2903 字 (约 12 分钟)
85

This article introduces Direct Preference Optimization (DPO) technology, which optimizes text generation by using rejection pairs from the model's own failures, significantly reducing text degradation rates. DPO is particularly effective in OCR tasks, as it can serve as a direct mitigation tool for specific failure modes without relying on subjective human judgments.

入选理由:DPO技术通过使用模型自身失败时产生的拒绝对来优化文本生成,显著减少了文本退化率。

FeaturedArticle#Direct Preference Optimization#OCR#text generation#model training中文
LiteParse is the best open-source, model-free document parser for AI agents.

Run it over over 50+ d...

LiteParse is the best open-source, model-free document parser for AI agents

Jerry Liu(@jerryjliu0)289 字 (约 2 分钟)
85

LiteParse is an open-source, model-free document parser that supports over 50 document types, quickly parses complex text layouts and tables, and extracts clean text in seconds, with lightweight OCR integrations.

入选理由:LiteParse 支持 50 多种文档类型,包括复杂的文本布局和表格。

FeaturedTweet#LiteParse#Document Parsing#Open Source#OCR英文
Excited to share that Qdrant will be speaking at the @MistralAI AI NOW Summit in Paris 🇫🇷

Chadha ...

Qdrant to Speak at MistralAI's AI NOW Summit in Paris

Qdrant(@qdrant_engine)106 字 (约 1 分钟)
45

Qdrant announces it will present a session combining OCR and semantic search for messy documents at MistralAI’s summit, though the post is merely an event announcement without technical depth.

入选理由:Qdrant 将在巴黎 AI NOW 峰会展示语义搜索与 OCR 结合的应用。

FeaturedTweet#Qdrant#MistralAI#Semantic Search#OCR#AI Conference英文

跨材料问答 · OCR

回答基于:OCR 相关 3 条材料
    0 / 500

    AI may generate inaccurate information. Please verify important content.