Perplexity 推出了新的人工智能上下文嵌入模型 pplx-embed-v2-context-9b-preview，该模型在 ConTEB 和 context-bench 基准测试中设定了新的最先进水平。该模型在查看整个文档的情况下对每个文档块进行编码，解决了传统分块方法的局限性。它通过从查询感知的上下文压缩模型中提炼相关性来实现这一点，并提供比 voyage-context-4 等模型更节省空间的向量表示。 AI
影响 在检索基准测试中设定了新的 SOTA，可能改进 RAG 系统并提供更高效的向量存储。
排序理由 Frontier-lab 模型发布，附带系统卡。
AI 生成摘要 · Google Gemini · 来自 7 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
View pplx-embed-v2-context-9b-preview on Hugging Face: https://t.co/EeqI2xCFOl
On ConTEB, the preview has the highest average nDCG@10 of the models tested, though not on every task. It also beats voyage-context-4 on chunk retrieval while using 8x less storage per vector: 1 KB (1024 dims, int8) vs 8 KB (2048, float32). https://t.co/PsyAwyU0Zo
pplx-embed-v2-context-9b-preview leads on Answer and Evidence retrieval at every cutoff. Document retrieval is closer, with our context v1 4B slightly ahead at Document@3 and Document@5. https://t.co/9DrBG7sxPp
context-bench is a benchmark for context-aware retrieval, created and privately held by @turbopuffer. Its queries, documents and capabilities are inspired by conversations with turbopuffer customers. It has 2,099 queries and 38,894 documents. We submitted for blind evaluation. h…
We overcome gold-chunk supervision limits by distilling relevance from our query-aware context compression model. It scores every document token against the query. Aggregated into chunk-level targets, those scores train the embedder to retrieve answer and supporting chunks. http…
Retrieval systems often split long documents into chunks, but this strips away the surrounding context. Contextual embedding models fix this by encoding the whole document once and pooling chunk vectors afterward. They are usually trained on one gold chunk per query.
We built a new way to train contextual embedding models, which encode each chunk of a document with the whole document in view. pplx-embed-v2-context-9b-preview sets a new state of the art on ConTEB and @turbopuffer's new, privately held context-bench. https://t.co/tkBUEjWVio h…