3 ms·
Funny result from the web demo. I'm well aware that it's an extremely small and, well, stupid, model, but even so: Query: HN Result: { "function_calls": [
by Tiberium 2mo ago
Funny result from the web demo. I'm well aware that it's an extremely small and, well, stupid, model, but even so:
Query: HN
Result:
{
"function_calls": [
{
"name": "lock_door",
"arguments": {
"door": "front door"
}
}
],
"reasoning": "User wants to lock the door. No specific door mentioned, so use 'front door' as default.",
"confidence": 0
}
I'd expect it to at least ignore (call no tools) for the queries that it doesn't understand. And it seems like it does do that, just not consistently.
- jszymborski 2mo agono, this is the appropriate response to hearing the words "HN" :P
- Schiendelman 2mo agoWas that the first message you sent it?
- yoavm 2mo agoThe website says the model is for "tool calling, device use, and structured extraction". Your example just doesn't seem to be very relevant. FWIW, it did a pretty good job for tool calling when I tried it, and I think it could be pretty nice to have this running on locally and integrate with Home Assistant.
- evmaki 2mo agoFalse positives are definitely relevant and worth measuring - natural language interfaces always have a discoverability problem, i.e., users not knowing what actions the system does and does not support. If the frontend of that system lacks the ability to reject unsupported commands, weird stuff happens. Nonetheless, this is very cool work! If I can offer a small suggestion to the team at Cactus, it would be to evaluate your releases on some usability criteria (including false positives). Any serious integrator or adopter of these models would want to have that information available.
- jdknezek 2mo ago> "confidence": 0 OP and the linked page talk about the confidence score and using it as an action threshold, so it looks like an appropriate total response to me.
- evmaki 2mo agoRight, but that's not the same thing as reporting a benchmark across a test set. It doesn't help me determine how well the model does across a decently-large sample size of commands. It doesn't tell me with what reliability the confidence will be below a given threshold when it should be, above that threshold when it should be, etc.
- HenryNdubuaku 2mo agoThanks, noted!
- curious_cat_163 2mo agoI think the test above is about tool calling... That's how I read it. The issue here is known as "out of distribution detection" in the old-timey classification world. I am not sure how a micro model will fundamentally solve it. Would love to understand what dannyw and team did there?
- derangedHorse 2mo agoHow did you draw an association between dannyw and Cactus? There are no 'Danny's on the list of GH contributors nor is there one named in the paper. Just curious.
- curious_cat_163 2mo agoAh, my bad. I may have misread it.
- planb 2mo agoThis is extremely impressive if it works. But on the other hand, if the number of cases where it works as expected is lower than what we could catch with a (old-Siri-style) heuristics based approach, and the rest fails in unpredictable ways, I'd prefer the dumb old "I did not understand that" response.
- petu 2mo ago"confidence": 0, so I guess you could threshold it
- justoneletter 2mo agoOnly if the confidences are calibrated, and they probably aren't. Any idea how the confidence is derived?
- hmokiguess 2mo agoyeah I got the same, almost like its biased heavily towards that as the 0 ranking -- my prompt was just the word 'potato'
- HenryNdubuaku 2mo agoThis is exactly why the confidence feature was introduced, the model knows when its wrong, we could hide that part and return a placeholder "sorry I only do function calls", would that be better or you prefer to see everything?
- tough 2mo agoconfidently stating that the confidence is right seems like a big claim
- plingbang 2mo agoI've got an identical output with the prompt "do not lock the door".
- nc55g3g 2mo ago[dead]
- kristjan 2mo agoIt seems to love locking doors. I was hoping this would dim the lights. > I'm hungover { "function_calls": [ { "name": "lock_door", "arguments": { "door": "front door" } } ], "reasoning": "User wants to lock the door. 'hungover' implies a security door. No specific door named, so use 'front door' as default.", "confidence": 0 }
- dvratil 2mo agoCannot unlock, though: Prompt: "Unlock my doors" { "function_calls": [], "reasoning": "No tool for unlocking or locking door exists. lock_door only performs lock/unlock actions, not unlock.", "confidence": 0.9492 }