Bài kiểm tra có liên kết hướng dẫn · Cứng Cấp độ
AI sáng tạo cho bài kiểm tra đánh giá tài liệu lần đầu
So sánh AI tổng quát với đánh giá được hỗ trợ bởi công nghệ, bao gồm xác thực, bảo mật và giám sát pháp lý đối với quy trình làm việc tài liệu.
Câu hỏi 1 của 8
How does technology-assisted review help prioritize a large document collection?
Xem trước mọi câu hỏi8 câu hỏi
Câu trả lời và giải thích xuất hiện khi bạn làm bài kiểm tra.
- How does technology-assisted review help prioritize a large document collection?
- Prioritize documents under a defined review protocol
- Determine the outcome of a lawsuit
- Replace all human coding
- Create new evidence absent from the record
- Why should an AI-generated summary be checked against source documents?
- Summaries always preserve every detail
- Source documents cannot be searched
- A fluent summary can omit a qualification or misstate content
- Legal teams should avoid identifying relevant passages
- Which measure can help assess a retrieval workflow’s missed-document risk?
- Number of pages per document
- Precision among retrieved documents, without reviewing material that was missed
- Average summary length
- Number of documents processed, regardless of relevant documents not found
- What should a legal review team define before model-assisted review?
- The final legal conclusion before examining documents
- Only the preferred generative model
- A prompt that avoids human review
- The objective, population, privilege process, and quality checks
- Why preserve document identifiers and source links?
- So findings can be traced back to specific records
- So reviewers cannot change a coding decision
- So a model can infer privilege automatically
- So the original document can be discarded
- Which concern is especially important when processing client documents?
- Model response speed only
- The number of clauses in a contract
- Confidentiality, access controls, retention, and vendor terms
- Whether the summary uses first-person language
- What does one overall accuracy score fail to show?
- Whether the system used a computer
- Which relevant documents were missed and how review thresholds behaved
- How many documents were in the corpus
- Whether the prompt was readable
- Who should make a privilege or production decision?
- The model based only on a confidence score
- The vendor without matter context
- The first reviewer who sees a summary
- Appropriately qualified legal reviewers under the governing process