Overview
Who this competition fits
适合已有中文 NLP 和任务型对话处理经验、想做槽值共指消解的参赛者。你需要能处理按词切分的多轮对话,并理解说话者、命名实体和成分句法等结构化输入及全局 span 索引。
Read the original official blurb
本评测重要关注任务型对话系统中的对话共指消解,即预测对话中所有提及同一槽值的指代。 Official tags: 对话共指消解. Sign-ups: 148.
Resources
What this competition gives you
Taken from the competition's official page.
Preparation
From registration to a first submission
- 01
中文自然语言处理
- 02
共指消解建模
- 03
任务型对话数据处理
- 04
JSON Lines 读写
- 05
span 索引标注
- 06
PyTorch 或飞桨
Before you commit: 核心难点是跨多轮对话追踪同一槽值的不同指称,并在多个候选实体间正确建立 cluster;这要求同时处理远距离上下文和多槽值指称。提交也需将预测结果整理为单个 UTF-8 JSON 文件中的 span-index cluster,并附完整代码与系统报告。
Source
How this page was assembled
Competition information is structured from the official page. Scores are platform estimates for decision support; official rules take precedence.
- Official competition page
- Baidu AI Studio
- Last checked
- Not recorded