This hits a sweet spot I think for conversations too. I've been playing (for quite a while) on trying to encapsulate long running conversations.

You have the overriding context, facts that don't change very often at all. The participants names, their backgrounds etc.

Then you have some very fine grained facts (what they ate for breakfast this morning) which might be useful right now, but are irrelevant outside of a general trend over the longer term.

When trying to reconstruct a conversation you really need to find the right balance without pulling in everything that has ever been discussed.

This definitely is worth further investigation.

▲

timwis 2 hours ago | parent | next [-]

Can you say more about how this applies to long-running conversations? I've been thinking about them as well, but can't write wrap my head around how this would be better than (or even different to) standard compaction.

▲

dominotw an hour ago | parent [-]

standard compactions doesnt really distinguish between long term vs short term ephemeral facts ?

	▲	timwis 37 minutes ago \| parent [-]
		Forgive me if I'm being naive, but can't you just tweak the compaction prompt to differentiate? Presumably that's what you would do in the separate prompt anyway, right?

▲

ewild 3 hours ago | parent | prev [-]

This sounds like we are trying to add an LSTM into a transformer

	▲	htrp 2 hours ago \| parent [-]
		Sepp would like a word