按年被提到
20222023 · 2 次2026
大模型与语言模型›词向量与预训练模型·arXiv:2212.09741 · cs.CL

One Embedder, Any Task: Instruction-Finetuned Text Embeddings

Hongjin Su, Weijia Shi, Jungo Kasai, Yizhong Wang, Yushi Hu, Mari Ostendorf, Wen-tau Yih, Noah A. Smith, Luke Zettlemoyer, Tao Yu · 2022
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2023-04-28 · 最近 2023-07-12
2不同的人
2活跃月份
2讨论串
83被引(OpenAlex)
arXiv 摘要(候选层不写导读)

We introduce INSTRUCTOR, a new method for computing text embeddings given task instructions: every text input is embedded together with instructions explaining the use case (e.g., task and domain descriptions). Unlike encoders from prior work that are more specialized, INSTRUCTOR is a single embedder that can generate text embeddings tailored to different downstream tasks and domains, without any further training. We first annotate instructions for 330 diverse tasks and train INSTRUCTOR on this multitask mixture with a contrastive loss. We evaluate INSTRUCTOR on 70 embedding evaluation tasks (66 of which are unseen during training), ranging from classification and information retrieval to semantic textual similarity and text generation evaluation. INSTRUCTOR, while having an order of magnitude fewer parameters than the previous best model, achieves state-of-the-art performance, with an average improvement of 3.4% compared to the previous best results on the 70 diverse datasets. Our analysis suggests that INSTRUCTOR is robust to changes in instructions, and that instruction finetuning mitigates the challenge of training a single model on diverse datasets. Our model, code, and data are available at https://instructor-embedding.github.io.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

没有评论直接链到它;下面是它被提交的讨论串里回复最多的评论(不计入统计)
讨论串评论lyjackal · 2023-04

“Instructor is at the top of the huggingface embedding benchmark leaderboard (https://huggingface.co/spaces/mteb/leaderboard), but I haven't heard much about it. In my testing it works pretty well relative to others as long as the prompt is crafted in the correct format.”

One Embedder, Any Task: Instruction-Finetuned Text Embeddings原评论 ↗
被提交为 story · 2 次