NLP-progress

On further review, there are more GitHub projects labeled with https://github.com/topics/text-summarization than "automatic-summarization"; e.g. awesome-text-summarization: https://github.com/icoxfog417/awesome-text-summarization and https://github.com/luopeixiang/awesome-text-summarization , which links to what look like relatively current benchmarks for SOTA performance in text summarization from the gh source repo of https://nlpprogress.com/ : https://github.com/sebastianruder/NLP-progress/blob/master/e...

I asked the (at the time) SOTA author to replace its BERT implementation with XLnet. He accepted and won 0.5% of accuracy on constituency parsing https://github.com/sebastianruder/NLP-progress/blob/master/e... which is actually a huge, it means 5% less errors. Which tasks remain to be tried? Most actually but an obvious one would be coreference resolution https://github.com/sebastianruder/NLP-progress/blob/master/e...

I work on arguably the most interesting of all machine learning tasks: semantic parsing (https://github.com/sebastianruder/NLP-progress/blob/master/e...) Such a tasks would benefit from a graph database. I'm currently going the mongoDB route but eventually a true graphDB would be a better fit

The opencog website is a great resource. Going directly to the specification is a bit too intimidating but here it is: https://wiki.opencog.org/w/CogPrime_Overview You might just begin by learning the list of NLP tasks and how good are the state of the art at it. The cognitive architecture that needs to be created to achieve AGI will one way or another be a composition of said tasks, which are the primitives. You can discover such a taxonomy here: https://github.com/sebastianruder/NLP-progress/blob/master/R... Also you might be interested by learning logic as a big task is to translate natural language into queryable, logical forms.

I'm not sure. I rarely have to do that explicitly in my head. Well I can't prove it but I strongly believe that our brains use part of speech too, unconsciously. Perhaps a model should learn to infer/guess them implicitly, from data. That's exactly what deep learning POS tagger do, they are far better than hard coded algorithms. SOTA has 97.96% of accuracy. In my opinion, having a world model (for common sense) and situational awareness (e.g. through sensor fusion, or from prior conversational history, or using some externally supplied conditioning) would be far more important. Haha you basically want a general intelligence (AGI), I want it too! And not enough persons works on "architecting" such a thing. Opencog may interest you a lot then. But the reality is many other "simpler" tasks are needed to make this happen. having a world mode

State of the art on most NLP tasks is now provided by deep learning, thanks to a host of techniques for dealing with both categorical and structurally complex data, such as LSTMs. Here's a great resource for keeping up with development in that area: https://github.com/sebastianruder/NLP-progress

As others have mentioned, I have no idea what any of this has to do with blockchain, or how it could even conceivably help with anything other than riding a hype train. That being said, would be curious to see how it holds up against the standard state of the art (e.g. https://github.com/sebastianruder/NLP-progress/blob/master/e...) as summarization (especially the abstractive kind) is very hard, but also has a surprising amount of useful applications

Pretrained models have enabled so much in CV, excited to see similar shifts in the language world. A great supplement is Sebastian’s NLP progress repo: https://github.com/sebastianruder/NLP-progress