Mastodon 🐘

Axel PichlerJust compared Claude-Sonnet-3.5 with OpenAI's o1 on a CLS task – classifying text inputs from US short stories with regard to focalization. Turns out, Sonnet doesn't recognize zero focalization and achieved an F1-score of 0.47, while o1 performed better with 0.69. Not bad - but problematic, as the hidden tokens of the optimizer (?) from o1 would be of particular interest.<a href="https://fedihum.org/tags/CLS" class="mention hashtag" rel="nofollow noopener" target="_blank">#CLS</a> <a href="https://fedihum.org/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://fedihum.org/tags/ClaudeSonnet" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeSonnet</a> <a href="https://fedihum.org/tags/OpenAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#OpenAI</a>'s_o1 <a href="https://fedihum.org/tags/TextClassification" class="mention hashtag" rel="nofollow noopener" target="_blank">#TextClassification</a> <a href="https://fedihum.org/tags/Focalization" class="mention hashtag" rel="nofollow noopener" target="_blank">#Focalization</a>

hexaheximalI've started making a text classification library in Go. It's VERY WIP, but I think that there could be some interesting results once I add things to it. (and maybe redesign the entirety of it if needed in the future)Source code: <a href="https://codeberg.org/autodni/silt" rel="nofollow noopener" translate="no" target="_blank">https://codeberg.org/autodni/silt</a><a href="https://mastodon.social/tags/Go" class="mention hashtag" rel="nofollow noopener" target="_blank">#Go</a> <a href="https://mastodon.social/tags/GoLang" class="mention hashtag" rel="nofollow noopener" target="_blank">#GoLang</a> <a href="https://mastodon.social/tags/Language" class="mention hashtag" rel="nofollow noopener" target="_blank">#Language</a> <a href="https://mastodon.social/tags/NLP" class="mention hashtag" rel="nofollow noopener" target="_blank">#NLP</a> <a href="https://mastodon.social/tags/Classification" class="mention hashtag" rel="nofollow noopener" target="_blank">#Classification</a> <a href="https://mastodon.social/tags/TextClassification" class="mention hashtag" rel="nofollow noopener" target="_blank">#TextClassification</a>

Data BloggerGenerative or Discriminative? How does Multinomial Naive Bayes work for text classification? 📚🐍🤔 <a href="https://mastodon.social/tags/NaiveBayes" class="mention hashtag" rel="nofollow noopener" target="_blank">#NaiveBayes</a> <a href="https://mastodon.social/tags/TextClassification" class="mention hashtag" rel="nofollow noopener" target="_blank">#TextClassification</a> <a href="https://mastodon.social/tags/DataScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#DataScience</a> <a href="https://towardsdatascience.com/multinomial-naive-bayes-classifier-c861311caff9" rel="nofollow noopener" translate="no" target="_blank">https://towardsdatascience.com/multinomial-naive-bayes-classifier-c861311caff9</a>

MT Group at FBKOur pick of the week by <a href="https://sigmoid.social/@apierg" class="u-url mention" rel="nofollow noopener" target="_blank">@apierg</a>: Tokpo et al., 2023 "How Far Can It Go?: On Intrinsic Gender Bias Mitigation for Text Classification"<a href="https://arxiv.org/abs/2301.12855v1" rel="nofollow noopener" target="_blank">https://arxiv.org/abs/2301.12855v1</a><a href="https://sigmoid.social/tags/NLProc" class="mention hashtag" rel="nofollow noopener" target="_blank">#NLProc</a> <a href="https://sigmoid.social/tags/Gender" class="mention hashtag" rel="nofollow noopener" target="_blank">#Gender</a> <a href="https://sigmoid.social/tags/GenderBias" class="mention hashtag" rel="nofollow noopener" target="_blank">#GenderBias</a> <a href="https://sigmoid.social/tags/Mitigation" class="mention hashtag" rel="nofollow noopener" target="_blank">#Mitigation</a> <a href="https://sigmoid.social/tags/TextClassification" class="mention hashtag" rel="nofollow noopener" target="_blank">#TextClassification</a>

Hiram Ringtldr; <a href="https://mstdn.social/tags/data" class="mention hashtag" rel="tag">#data</a> <a href="https://mstdn.social/tags/augmentation" class="mention hashtag" rel="tag">#augmentation</a> in <a href="https://mstdn.social/tags/NLProc" class="mention hashtag" rel="tag">#NLProc</a> degrades <a href="https://mstdn.social/tags/textClassification" class="mention hashtag" rel="tag">#textClassification</a> performance in most cases. <a href="https://aclanthology.org/2022.insights-1.12.pdf" target="_blank" rel="nofollow noopener" translate="no">https://aclanthology.org/2022.insights-1.12.pdf</a>Well that was fun. Just spent last night experimenting with <a href="https://mstdn.social/tags/dataAugmentation" class="mention hashtag" rel="tag">#dataAugmentation</a> (using <a href="https://mstdn.social/tags/flan" class="mention hashtag" rel="tag">#flan</a>-T5 for paraphrasing) and in wondering why it seems to degrade <a href="https://mstdn.social/tags/textClassification" class="mention hashtag" rel="tag">#textClassification</a> performance I came across this great paper essentially saying the same thing. I guess I’ll revisit this in a year or so when there are better language models. <a href="https://a.gup.pe/u/linguistics" class="u-url mention">@linguistics</a> <a href="https://sigmoid.social/about/more?instance_actor=true" class="u-url mention">@sigmoid.social</a>

jfrolesAnyone use a machine learning api in their app? I had trouble finding a dead simple, cost effective one and ended up creating my own for <a href="https://techhub.social/tags/textclassification" class="mention hashtag" rel="nofollow noopener" target="_blank">#textclassification</a> . Curious if other people would get value out of it.<a href="https://techhub.social/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#ai</a> <a href="https://techhub.social/tags/serverless" class="mention hashtag" rel="nofollow noopener" target="_blank">#serverless</a> <a href="https://techhub.social/tags/buildinpublic" class="mention hashtag" rel="nofollow noopener" target="_blank">#buildinpublic</a>

Fabian HoppeHi everybody, <a href="https://mathstodon.xyz/tags/introduction" class="mention hashtag" rel="nofollow noopener" target="_blank">#introduction</a> I'm a PhD student at <a href="https://mathstodon.xyz/tags/AIFB" class="mention hashtag" rel="nofollow noopener" target="_blank">#AIFB</a>/KIT and part of the <a href="https://mathstodon.xyz/tags/FIZKarlsruhe" class="mention hashtag" rel="nofollow noopener" target="_blank">#FIZKarlsruhe</a> <a href="https://sigmoid.social/@fizise" class="u-url mention" rel="nofollow noopener" target="_blank">@fizise</a> research group. On this account you will find mostly toots about my research in the areas of <a href="https://mathstodon.xyz/tags/TextClassification" class="mention hashtag" rel="nofollow noopener" target="_blank">#TextClassification</a>, <a href="https://mathstodon.xyz/tags/ZSL" class="mention hashtag" rel="nofollow noopener" target="_blank">#ZSL</a>, <a href="https://mathstodon.xyz/tags/ZSTC" class="mention hashtag" rel="nofollow noopener" target="_blank">#ZSTC</a> and <a href="https://mathstodon.xyz/tags/KnowledgeGraph" class="mention hashtag" rel="nofollow noopener" target="_blank">#KnowledgeGraph</a>. I'm especially interested in everything connecting subsymbolic and symbolic <a href="https://mathstodon.xyz/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a>. Aside from this I might toot about <a href="https://mathstodon.xyz/tags/running" class="mention hashtag" rel="nofollow noopener" target="_blank">#running</a> and <a href="https://mathstodon.xyz/tags/coffee" class="mention hashtag" rel="nofollow noopener" target="_blank">#coffee</a>.

Recent searches

Search options

Administered by:

Server stats:

#TextClassification