4 ms·

Books that operate in a narrative/temporal style are hard to manage, because statements of fact are mutative. Consider the story: "Justin is hungry. Justin ea

Books that operate in a narrative/temporal style are hard to manage, because statements of fact are mutative. Consider the story: "Justin is hungry. Justin eats dinner. Justin is not hungry." You ask the chatbot "Is Justin hungry?". There is a temporal aspect to this question that is hard for simple systems that are just embedding facts into a vector DB (or similar techniques) to reconcile.

eddsh1994 4y ago
Is that how LLMs work?
lgas 4y ago
I asked ChatGPT: Me: Consider the story: "Justin is hungry. Justin eats dinner. Justin is not hungry." Is Justin hungry? ChatGPT: No, Justin is not hungry after eating dinner. I'm not sure that it's that big of a problem.
- mnkm 4y ago
  It has more to do with people's expectations of fiction books being different for this format. With non-fiction it's more straight forward. Simple Q&A.
- gamegoblin 4y ago
  The example was to just illustrate the general problem. Think of ingesting a whole novel that takes place over a few years. The whole novel doesn't fit into GPT's context window (which is only a page or two of text). So you have to extract individual statements of fact and index over them (e.g. with semantic indexing, or many other techniques). It's tricky to deal with cases where the state of something changes many times over the course of the years in the novel. Imagine you ingest the whole Harry Potter series. You ask the chatbot "How old is Harry Potter?". The answer to the question depends on which part of the story you are talking about. "Does Harry know the foobaricus spell?" The answer depends on which part of the story you are talking about. Whereas for a non-fiction book typically does not contain these temporally changing aspects. In a book about astronomy, Mars is the 4th planet from the sun in chapter 1, and in chapter 10.
  - lgas 4y ago
    > Think of ingesting a whole novel that takes place over a few years. I did exactly that with Asimov's Let's Get Together using https://github.com/jerryjliu/gpt_index https://github.com/jerryjliu/gpt_index. It's a short story that's only 8,846 words, so it's not quite a novel, much less the whole of the Harry Potter series, but it was able to answer questions that required information from different parts of the text all at the same time. It requires multiple passes of incremental summarization so it is of course much slower than making a single call to the model, but I stand by my assertion that these things just aren't much problem in practice. They are only a problem if you're trying to paste them into ChatGPT or the GPT-3 playground window or something like that. People are solving the problems with building these systems in the real world almost as fast as the problems arise in the first place.
  - alexpotato 4y ago
    I would imagine that the "attention" phase of the LLMs could get longer over time as more resources are dedicated to them. e.g. we are seeing the equivalent of movies that are 5 minutes long b/c they were hand animated. Once we move to computer animated movies, it becomes a lot easier to generate an entire film.
    - gamegoblin 4y ago
      I agree they will get longer. ChatGPT (GPT3.5) is 2x larger than GPT3. 8192 tokens vs 4096. The problem is that in the existing transformer architecture, the complexity of this is O(N^2). Making the context window 10x larger involves 100x more memory and compute. We'll either need a new architecture that improves upon the basic transformer, or just wait for Moore's law to paper over the problem for the scales we care about. In the short term, you can also use the basic transformer with a combination of other techniques to try to find the relevant things to put into the context window. For instance, I ask "Does Harry Potter know the foobaricus spell?" and then the external system does a more traditional search technique to find all sentences relevant to the query in the novels, maybe a few paragraph summary of each novel, etc, then feeds that ~1 page worth of data to GPT to then answer the question.
anigbrowl 4y ago
Don't be so sure about that. Pick some fictional work you like and ask ChatGPT to drill down into the factors that constrain and motivate the characters and be surprised. You might need to go through a few exchanges to filter out inaccurate or misremembered summaries. LLMs also lack (for now) an appreciation of what givens have large significance for humans because they are so obvious, eg discoursing and speculating on the problems a character encounters by making inferences about personality and life situation, but neglecting to mention at first that the person is blind. For a human this would be a fundamental distinguishing factor, whereas to an LLM it appears as just another factoid unless its specifically asked to evaluate its own contextual frame.

eddsh1994 4y ago

Is that how LLMs work?

lgas 4y ago

I asked ChatGPT: Me: Consider the story: "Justin is hungry. Justin eats dinner. Justin is not hungry." Is Justin hungry? ChatGPT: No, Justin is not hungry after eating dinner. I'm not sure that it's that big of a problem.

mnkm 4y ago
It has more to do with people's expectations of fiction books being different for this format. With non-fiction it's more straight forward. Simple Q&A.
gamegoblin 4y ago
The example was to just illustrate the general problem. Think of ingesting a whole novel that takes place over a few years. The whole novel doesn't fit into GPT's context window (which is only a page or two of text). So you have to extract individual statements of fact and index over them (e.g. with semantic indexing, or many other techniques). It's tricky to deal with cases where the state of something changes many times over the course of the years in the novel. Imagine you ingest the whole Harry Potter series. You ask the chatbot "How old is Harry Potter?". The answer to the question depends on which part of the story you are talking about. "Does Harry know the foobaricus spell?" The answer depends on which part of the story you are talking about. Whereas for a non-fiction book typically does not contain these temporally changing aspects. In a book about astronomy, Mars is the 4th planet from the sun in chapter 1, and in chapter 10.
- lgas 4y ago
  > Think of ingesting a whole novel that takes place over a few years. I did exactly that with Asimov's Let's Get Together using https://github.com/jerryjliu/gpt_index https://github.com/jerryjliu/gpt_index. It's a short story that's only 8,846 words, so it's not quite a novel, much less the whole of the Harry Potter series, but it was able to answer questions that required information from different parts of the text all at the same time. It requires multiple passes of incremental summarization so it is of course much slower than making a single call to the model, but I stand by my assertion that these things just aren't much problem in practice. They are only a problem if you're trying to paste them into ChatGPT or the GPT-3 playground window or something like that. People are solving the problems with building these systems in the real world almost as fast as the problems arise in the first place.
- alexpotato 4y ago
  I would imagine that the "attention" phase of the LLMs could get longer over time as more resources are dedicated to them. e.g. we are seeing the equivalent of movies that are 5 minutes long b/c they were hand animated. Once we move to computer animated movies, it becomes a lot easier to generate an entire film.
  - gamegoblin 4y ago
    I agree they will get longer. ChatGPT (GPT3.5) is 2x larger than GPT3. 8192 tokens vs 4096. The problem is that in the existing transformer architecture, the complexity of this is O(N^2). Making the context window 10x larger involves 100x more memory and compute. We'll either need a new architecture that improves upon the basic transformer, or just wait for Moore's law to paper over the problem for the scales we care about. In the short term, you can also use the basic transformer with a combination of other techniques to try to find the relevant things to put into the context window. For instance, I ask "Does Harry Potter know the foobaricus spell?" and then the external system does a more traditional search technique to find all sentences relevant to the query in the novels, maybe a few paragraph summary of each novel, etc, then feeds that ~1 page worth of data to GPT to then answer the question.

anigbrowl 4y ago

Don't be so sure about that. Pick some fictional work you like and ask ChatGPT to drill down into the factors that constrain and motivate the characters and be surprised. You might need to go through a few exchanges to filter out inaccurate or misremembered summaries. LLMs also lack (for now) an appreciation of what givens have large significance for humans because they are so obvious, eg discoursing and speculating on the problems a character encounters by making inferences about personality and life situation, but neglecting to mention at first that the person is blind. For a human this would be a fundamental distinguishing factor, whereas to an LLM it appears as just another factoid unless its specifically asked to evaluate its own contextual frame.