4 ms·
Situation: people try to use these predictive text chatbots as search engines. Problem: LLMs are not search engines. They extrapolate, interpolate, and approxi
by wizzwizz4 3y ago
Situation: people try to use these predictive text chatbots as search engines.
Problem: LLMs are not search engines. They extrapolate, interpolate, and approximate (so-called “hallucinations”) so they can always produce somewhat-plausible text completions.
Solution: Create a search engine so good at returning relevant results that even an LLM can make use of it… then go to significant lengths to plug that search engine into the LLM, preventing people from reading the search results directly.
Why not simply give people access to the search engine‽ People know how to use search engines!! This is the fifth time I've seen an article like this, and I'm still… baffled. It's https://xkcd.com/2021/ https://xkcd.com/2021/ all over again.
- omeze 3y agoYea… my opinion on this is that startups are attempting to force a market for chat bots, instead of accepting that LLM embeddings are best utilized as a search feature, not new product surface area
- woeirua 3y agoThey really want agents to work, and they just don’t yet.
- CSMastermind 3y agoIMO the most obvious place LLMs will be used is for interfaces. They enable voice-based interfaces to be practical for normal users for the first time since you really can talk to them in a convincing way. Translating user input into a set of well-defined commands seems like a better use than searching data to me.
- potatoman22 3y agoWell, search engines and my ability to query them aren't good enough for highly specific or poorly worded questions yet. LLMs are sometimes better in this space.
- jwells89 3y agoIndeed, a frequent issue with search engines is knowing the right terms plug into them, particularly when researching a topic beyond one's scope of knowledge. Half the power of LLMs as they currently exist is that they can often extract the intention of the user's question in a way that search engines usually can't, allowing them to provide a more useful answer or at least point the user in the right direction. Perhaps it would make sense for search engines to utilize LLMs to perform this query extraction and suggest more appropriate search terms, engaging conversational interaction only if the suggestions are wrong and the LLM requires further clarification.
- lalopalota 3y agoI agree. Use the LLM to construct a better search query, then return those results and the query to the user. The user can modify the "optimized" query and repeat until a usefull answer is obtained.
- worrycue 3y ago> Half the power of LLMs as they currently exist is that they can often extract the intention of the user's question in a way that search engines usually can't I don’t know about that. Google is pretty good at including “similar” questions that others have asked to what I have queried and often that’s exactly what I needed.
- jwells89 3y agoIn my experience, Google's ability to suggest "similar" queries is often limited if I don't know the terminology associated with the subject in question. It's decent if you're already in the ballpark, but to extend the analogy if you're stuck trying to figure out where the park is in the first place it's much more hit or miss.
- worrycue 3y agoI “fish” with google when I don’t know the terminology is what I’m saying. With some luck Google will have correct question on the first page of the search results.
- wmf 3y agoYeah, this "fix" just shifts the problem to the vector DB and its embedding algorithm. People keep forgetting to mention that embeddings aren't 100% accurate either. The net accuracy may be better but it's not magic.
- signatoremo 3y agoI’m sure you’ve heard of Google-fu. Someone is better at searching than others. I believe they propose that LLM can be superior at producing search queries. Whether they succeed, that remains to be seen, but the idea isn’t that baffling.
- zirgs 3y agoWhy not show the "raw" search results as well?