Tag
Shuo Yang and team release FlashLib, a GPU library that accelerates classical ML operators like KMeans, KNN, HDBSCAN, PCA, and t-SNE, claiming speedups up to 208x.
This paper presents a fully automated pipeline that transforms court decisions into legal commentaries by extracting, clustering, and summarizing paragraph-level chunks using LLMs, evaluated on German civil code cases.
This position paper argues that current uncertainty quantification methods for large language models are essentially unsupervised clustering, measuring internal consistency rather than external correctness, and therefore fail to detect confident hallucinations. The authors advocate for a paradigm shift to ground uncertainty in objective truth.
This model maps sentences to 384-dimensional vectors for tasks like clustering and semantic search, fine-tuned on 1B sentence pairs using contrastive learning. It is part of the sentence-transformers library and can be used with Hugging Face Transformers.