undefined | Better HN

0 pointsverdverm6mo ago0 comments

> the trade-off is that instead of returning links it returns a generated soup of words that are semantically close to your "query".

I get links in my responses from Gemini. I would also not describe the response as soup, the answers are often quite specific and in the terms of my prompt instead of the inputs (developer queries are a prime example)

0 comments

quectophoton6mo ago

I call them a "soup" because AFAIK there's no intent behind them:

I'll stop calling them a soup when the part that generates a human-readable response is completely separate from the knowledge/information part; when an untrained program can respond with "I don't know" due to deliberate (/debuggable) mapping of lack of data to a minimal subset of language rules and words that are encoded in the program, rather than having "I don't know" be a series of tokens generated from the training data.

verdvermOP6mo ago

Those are called Agents and already exists today. I've been prompted for more information when the agent realized it didn't have all the context it needs

quectophoton6mo ago

Don't agents still depend on LLMs to produce a human-readable response, rather than as a source of information/knowledge? And aren't they still vulnerable to prompt injection attacks, due to being unable to separate the information/knowledge part vs the prompt, because their prompt "parsing" is coupled to an LLM?

1 more reply

j / k navigate · click thread line to collapse

0 pointsverdverm6mo ago0 comments

> the trade-off is that instead of returning links it returns a generated soup of words that are semantically close to your "query".

0 comments

quectophoton6mo ago

I call them a "soup" because AFAIK there's no intent behind them:

verdvermOP6mo ago

Those are called Agents and already exists today. I've been prompted for more information when the agent realized it didn't have all the context it needs

quectophoton6mo ago

1 more reply

j / k navigate · click thread line to collapse