Viewing profile — lgessler
lgessler
HN member- Joined
- Tue, Sep 29, 2015, 6:26 AM UTC
- HN karma
- 1,712
- Public activity
- 435 items
- HN profile
- View on Hacker News ↗
About lgessler
Recent public activity
-
comment
Comment #49198024
I'm confused. I think you're confusing Gemini and Gemma. Gemini is Google's frontier offering which is closed-weight, API-only, like most frontier models. Gemma is Google's open-we…
- comment
-
comment
Comment #48949319
You're saying it's important to have up-to-date facts stored in parametric knowledge? It seems to me like that's grown less and less important as agentic capabilities have grown. E…
-
comment
Comment #48782799
Accuracy isn't a meaningful metric here without reference to a specific task.
-
comment
Comment #48557171
Not that this really takes away from the substance of the article, but the first two paragraphs are giving heavy Claude smell. Semicolons, em dashes, "That sequencing matters"... I…
-
comment
Comment #48392299
Raphaël Millière has a very useful term for this kind of vacuous dismissal, the redescription fallacy ( https://arxiv.org/pdf/2401.03910 , page 9): > Recent debates have been cloud…
-
comment
Comment #47868161
I'll be really interested to hear qualitative reports of how this model works out in practice. I just can't believe that a model this small is actually as good as Opus, which is ru…
-
comment
Comment #47646092
Is Java or Haskell any closer to human language?
-
comment
Comment #47416925
Has everyone always nailed their implementation of every program on the first try? Of course not. Probably what happens most times is you first complete something that sorta works …
-
comment
Comment #47383111
I know this is mostly about keyword substitution but it still tickles me that you still write f(x) in this language and not (x)f given that Korean is SOV but I guess that's just ho…
-
comment
Comment #46984011
Let's be real here, regardless of what Boris thinks, this decision is not in his hands.
- comment
- comment
-
comment
Comment #45076319
Novels are fictional too. So long as they're not taken too literally, archetypes can be helpful mental prompts.
-
comment
Comment #44966516
If you're really just doing traditional NER (identifying non-overlapping spans of tokens which refer to named entities) then you're probably better off using encoder-only (e.g. htt…
-
comment
Comment #44887204
I recommend having a look at 16.3 onward here if you're curious about this: https://web.stanford.edu/~jurafsky/slp3/16.pdf I'm not familiar with Whisper in particular, but typicall…
- comment
-
comment
Comment #44527453
In my (poor) understanding, this can depend on hardware details. What are you running your models on? I haven't paid close attention to this with LLMs, but I've tried very hard to …
- comment
- comment
-
comment
Comment #43640148
Sure, this is a common sentiment, and one that works for some courses. But for others (introductory programming, say) I have a really hard time imagining an assignment that could n…
- comment
-
comment
Comment #43639035
I'm a professor at an R1 university teaching mostly graduate-level courses with substantive Python programming components. On the one hand, I've caught some students red handed (Ch…
-
comment
Comment #42662700
> Fermentation is completely stopped in a regular fridge, you need higher temperature for fermentation. My understanding as a hobbyist brewer and fermenter is that this is not true…
-
comment
Comment #42591408
I think this is both a harmful and irrational attitude. Why focus on some trivial mechanical errors and disparage the authors for it instead of the thing that is much more importan…