Eric Zhù·Jan 8Two ClaudesClaude Monet painted the same haystack over and over — at dawn, at dusk, in snow, in summer. He started late in the year of 1890 and kept…
InTDS ArchivebyEric Zhù·Dec 1, 2023Please Use Streaming Workload to Benchmark Vector DatabasesWhy static workload is insufficient and what I learned by comparing HNSWLIB and DiskANN using streaming workloadA response icon1A response icon1
InTDS ArchivebyEric Zhù·Aug 18, 2023Finding Needles in a Haystack — Search Indexes for Jaccard SimilarityFrom basic concepts to exact and approximate indexesA response icon1A response icon1
Eric Zhù·May 10, 2023GPT-4’s Maze Navigation: A Deep Dive into ReAct Agent and LLM’s ThoughtsIn this post I dissect a GPT-4 ReAct Agent capable of navigating a maze, and reveal GPT-4’s thoughts and built-in skills.A response icon3A response icon3
Eric Zhù·Mar 18, 2023Human-Aligned Text-to-SQL EvaluationIn my last post about Text-to-SQL using GPT-3.5, I pointed out the issue with existing benchmark’s evaluation metric: it rejects perfectly…
Eric Zhù·Mar 7, 2023What is Coming Next for Text-to-SQLText-to-SQL is a natural language processing (NLP) task that involves converting natural language questions into SQL queries that can be…
