News for hackers. Stuff that matters.
Re:Text Summarization with TensorFlow by peterjliu on Wednesday August 24, 2016 @04:27PM (#12354991)
Author of post here. I'd say most of the examples generated from the best model were good. However we chose examples that were not too gruesome, as news can be :)

We encourage you to try the code and see for yourself.

Re:Text Summarization with TensorFlow by probably_wrong on Wednesday August 24, 2016 @05:38PM (#12355445)
How does the model deal with dangling anaphora[1]? I wrote a summarizer for Spanish following a recent paper as a side project, and it looks as if I'll need a month of work to solve the issue.

[1] That is, the problem of selecting a sentence such as "He approved the motion" and then realising that "he" is now undefined.

Re:Text Summarization with TensorFlow by peterjliu on Wednesday August 24, 2016 @05:46PM (#12355512)
We're not "selecting" sentences as an extractive summarizer might. The sentences are generated.

As for how does the model deal with co-reference? There's no special logic for that.

Re:Text Summarization with TensorFlow by wutbrodo on Thursday August 25, 2016 @04:20AM (#12357575)
Wouldn't it suffice to do a coreference pass before extracting sentences? Obviously you'll compound coref errors with the errors in your main logic, but that seems somewhat unavoidable.
Re:Text Summarization with TensorFlow by mark_l_watson on Thursday August 25, 2016 @08:54AM (#12358598)
I am working on this in my kbsportal.com NLP demo. With accurate coreference substitutions (eg., substituting a previous NP like 'San Francisco' for 'there' in a later sentence, substituting full previously mentioned names for pronouns, etc.) extractive summarization should provide better results, and my intuition is that this preprocessing should help abstractive summarization also.
Re:Text Summarization with TensorFlow by flamedoge on Thursday August 25, 2016 @01:41AM (#12357138)
That is inter-sentence logic? Even humans have trouble with such ambiguity for certain cases.
Re:Text Summarization with TensorFlow by wintom on Wednesday August 24, 2016 @05:27PM (#12355381)
In the post you mentioned that

>>"In those tasks training from scratch with this model architecture does not do as well as some other techniques we're researching, but it serves as a baseline."

Can you elaborate a little on that? Is the training the problem or is the model just not good at longer texts?

Re:Text Summarization with TensorFlow by elyase on Wednesday August 24, 2016 @06:21PM (#12355702)
Any chance some trained model will be released?
Re:Text Summarization with TensorFlow by plusepsilon on Wednesday August 24, 2016 @10:37PM (#12356652)
Any hints on how to integrate the whole document for summarization? ;)

I've seen copynet, where you do seq2seq but also have a copy mechanism to copy rare words from the source sentence to the target sentence.

Re:Text Summarization with TensorFlow by fowlerpower on Wednesday August 24, 2016 @07:30PM (#12355999)
Is it hard to get the code up and running on Google Cloud? Does TensorFlow come as a service?