Evaluation |https://yzm1205.github.io/tags/evaluation/EvaluationHugoBlox Kit (https://hugoblox.com)en-usFri, 01 Nov 2024 00:00:00 +0000https://yzm1205.github.io/media/icon_hu_1c0e9cb08cfb822a.pngEvaluationhttps://yzm1205.github.io/tags/evaluation/ALIGN-SIM Resourceshttps://yzm1205.github.io/projects/align-sim/Fri, 01 Nov 2024 00:00:00 +0000https://yzm1205.github.io/projects/align-sim/<p><strong>Auburn University</strong></p> <ul> <li>Released software and data for task-free evaluation of sentence embeddings using five semantic similarity alignment criteria.</li> <li>Evaluated 13 classical and LLM-induced encoders and demonstrated that strong downstream benchmark performance can coexist with failures on intuitive semantic alignment tests.</li> </ul>ALIGN-SIM: A Task-Free Test Bed for Evaluating and Interpreting Sentence Embeddings through Semantic Similarity Alignmenthttps://yzm1205.github.io/publications/align-sim/Fri, 01 Nov 2024 00:00:00 +0000https://yzm1205.github.io/publications/align-sim/