[HN Gopher] Grounding AI in reality with a little help from Data...
       ___________________________________________________________________
        
       Grounding AI in reality with a little help from Data Commons
        
       Author : throwaway888abc
       Score  : 89 points
       Date   : 2024-09-13 20:41 UTC (1 days ago)
        
 (HTM) web link (research.google)
 (TXT) w3m dump (research.google)
        
       | westurner wrote:
       | > _Retrieval Interleaved Generation (RIG)_ : _This approach fine-
       | tunes Gemma 2 to identify statistics within its responses and
       | annotate them with a call to Data Commons, including a relevant
       | query and the model 's initial answer for comparison. Think of it
       | as the model double-checking its work against a trusted source._
       | 
       | > [...] _Trade-offs of the RAG approach_ : [...] _In addition,
       | the effectiveness of grounding depends on the quality of the
       | generated queries to Data Commons._
        
         | Groxx wrote:
         | Gotta say, this kinda feels like "giving it correct data didn't
         | work, so _what if we did that twice?_ ".
         | 
         | Like, two layers of duct tape are better than one.
         | 
         | Seems reasonable and I can believe it helps, just also seems
         | like it doesn't do much to improve confidence in the system as
         | a whole. Particularly since they're basically asking it to find
         | things worth checking, then have it write the checking query,
         | and have it interpret the results. When it's the thing that
         | screwed it up in the first place.
        
           | vineyardmike wrote:
           | eh, I think this is pretty reasonable and not a "hack". It
           | matches what we do as people. I think there probably needs to
           | be research into how to tell it when it doesn't know
           | something, however.
           | 
           | I think if you remember that LLMs are not databases, but they
           | do contain a super lossy-compressed version of (it's
           | training) knowledge, this feels less like a hack. If you ask
           | someone, "who won the World Cup in 2000?", they may say "I
           | think it was X, but let me google it first". That person
           | isn't screwed up, using tools isn't a failure.
           | 
           | If the context is a work setting, or somewhere that is data-
           | centric, it totally makes sense to check it. Like a Chat Bot
           | for a store, or company that is helping someone troubleshoot
           | or research. Anything where it really obvious answers that
           | are easy to learn from volumes of data ("what company makes
           | the corolla?"), probably don't need fact checking as often,
           | but why not have the system check its work?
           | 
           | Meanwhile, programming, writing prose, etc are not things you
           | generally fact-check mid-way, and are things that can be
           | "learned" well from statistical volume. Most programmers can
           | get "pretty good" syntax on first try, and any dedicated
           | syntax tool will get to basically 100%, and the same makes
           | sense for an LLM.
        
             | exe34 wrote:
             | I think a better way would be to just use it as a text to
             | search interface in the first place?
        
             | westurner wrote:
             | This is similar to the difference between data dredging and
             | scientific hypothesis testing.
             | 
             | 'But what bias did we infer with [LLM knowledgebase]
             | background research prior to formulating a hypothesis, and
             | who measured?'
             | 
             | There are various methods of Inference: Inductive,
             | Deductive, and Abductive
             | 
             | What are the limits of Null Hypothesis methods of
             | scientific inquiry?
        
           | taneq wrote:
           | > Like, two layers of duct tape are better than one.
           | 
           | Uh... they _are_? They 're not better than a properly specc'd
           | fastener installed at appropriately engineered mounting
           | points but still better than one layer of duct tape, let
           | alone none.
        
       | zrank wrote:
       | So, essentially, you can Google the answer in the first place,
       | click on your own trusted sources and compare them. All without
       | using a language model.
       | 
       | Or buy an encyclopedia ...
        
         | gaogao wrote:
         | There's an encylopedia you can't buy called Wikipedia. It also
         | has its own structured auxillary database, WikiData. I wrote up
         | roughly what this article is doing about a year and a half ago
         | - https://friend.computer/jekyll/update/2023/04/30/wikidata-
         | ll...
        
         | fassssst wrote:
         | Google uses language models.
        
       | mark_l_watson wrote:
       | I was fortunate to be hired as a contractor 11 years ago to work
       | on an internal Google Knowledge Graph application. Google is just
       | one of many large companies to utilize one huge graph to localize
       | information from many sources.
       | 
       | I bought into TBL's Semantic Web ideas (and I cover 'lower case'
       | semantic web topics in a few of my books). I think it is a shame
       | that publicly accessible world knowledge graphs never really took
       | off, but at least Google's Data Commons is available for free for
       | non-commercial, educational, and research uses.
        
         | ramraj07 wrote:
         | Could you provide any insight into why KGs work well in such
         | places? Like a contrived example maybe?
        
           | mark_l_watson wrote:
           | Well, from public information: Meta has a huge social graph
           | that helps support their social media businesses, and Google
           | has a wealth of real world knowledge. These graphs are, I
           | think, optimized for super fast 1 millisecond level query
           | latencies, and not a rich query language (for example, not
           | something like SPARQL).
           | 
           | I have just been looking at the Data Commons data sets, and I
           | think I will add a fun example to my live Common Lisp eBook
           | (and/or my Racket live eBook).
        
         | zozbot234 wrote:
         | > publicly accessible world knowledge graphs never took off
         | 
         | Huh? What's wrong with Wikidata and the Linked Open Data cloud?
         | These seem quite real to me.
        
       | openrisk wrote:
       | The public / non-government sector (especially in Europe) has
       | been quite keen for decades in (linked) open data, knowledge
       | graphs and associated technologies. Yet applications, tools and,
       | ultimately usability, awareness and adoption have been lagging.
       | 
       | In this sense this project offers a remarkable, albeit implicit,
       | endorsement of that broader open data space, as it comes from a
       | major private sector entity _and_ links with the hot LLM
       | technology of the day.
       | 
       | At high level though, this design seems to violate the "bitter
       | lesson" gospel [1].
       | 
       | > 1) AI researchers have often tried to build knowledge into
       | their agents
       | 
       | Which is at the same refreshing (as there is something very
       | incomplete and self-defeating in the "scaling" hypothesis) and
       | hints at the difficulties of meaningfully integrating very
       | heterogeneous sources and representations of information.
       | 
       | [1] http://www.incompleteideas.net/IncIdeas/BitterLesson.html
        
       | amelius wrote:
       | Information isn't the only problem. Another problem is the
       | correct application of logic.
        
       ___________________________________________________________________
       (page generated 2024-09-14 23:01 UTC)