Experience
Other similar profiles
Explore more posts
-
Matthew Farmer PhD, RN
moving beyond NLP-focused quantitative classification to stronger, qualitative-informed benchmarks for LLMs is substantially better at helping understand the growing human-AI interactions. Novel and foundational work, Kriti Aggarwal and team. I hope this inspires more mixed methods benchmarking in other domains.
-
Qihang（Tom） Cheng
Been trying out a semantic chunking strategy recently. The logic is to check if the max_sim between a new sentence and the current chunk is greater than the min_sim within the chunk. If so, merge; otherwise, create a new one. But in practice, I found that as soon as one irrelevant sentence gets into the chunk, the min_sim plummets, and then it pretty much stops creating new chunks. Also, just looking at max_sim can chain two unrelated topics together with just a few transitional sentences, which is a real headache. Performance-wise, the O(m²) complexity is just unusable. So, I changed my approach: for small chunks, I only use a max_sim threshold. Once the chunk gets larger (e.g., >5 sentences), I then bring in quantile-based constraints. This change really helped mitigate the chain-linking issue. For performance, you still have to use incremental calculations or approximate quantiles, or it's just too slow.
-
Ted Kyi
Another great new educational series is starting at San Diego Machine Learning. With the explosion of AI, many people have been asking how to jump start their ML skills. SDML is responding with a weekly program based on the Hands-On Machine Learning book by Aurelien Geron. This free series will be available both in person and online on Saturdays, starting August 2nd. There are no prerequisites, and people who complete the program will learn all of the core ML skills needed to excel. https://lnkd.in/gB_3N2S4 #ai #ml #datascience #machinelearning
7 Comments
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content