Nino
@ninohardt.bsky.social
Marketing scientist, market researcher, choice modeler.
web: ninohardt.com
web: ninohardt.com
Waitlist. How will it compare to chat.inceptionlabs.ai?
Mercury Coder
Mercury Coder
chat.inceptionlabs.ai
May 21, 2025 at 5:57 PM
Waitlist. How will it compare to chat.inceptionlabs.ai?
Parallels to reward hacking issue?
April 2, 2025 at 3:45 PM
Parallels to reward hacking issue?
Based on the image, it works
March 19, 2025 at 7:04 PM
Based on the image, it works
@philschmid.bsky.social saw you tweet (_philschmid/status/1902403286902178000) saying "Understands the text and visuals like Charts in the PDF", but I have doubts.
March 19, 2025 at 7:03 PM
@philschmid.bsky.social saw you tweet (_philschmid/status/1902403286902178000) saying "Understands the text and visuals like Charts in the PDF", but I have doubts.
@rao2z.bsky.social introduced the term "prompt augmentation". The holy grail would be a skolem function that finds the best prompt augmentation for an optimal output, whatever it may look like (it might not even make sense to humans).
March 9, 2025 at 4:27 PM
@rao2z.bsky.social introduced the term "prompt augmentation". The holy grail would be a skolem function that finds the best prompt augmentation for an optimal output, whatever it may look like (it might not even make sense to humans).
Turns out it isn't quite so straightforward: arxiv.org/abs/2502.21098
Re-evaluating Theory of Mind evaluation in large language models
The question of whether large language models (LLMs) possess Theory of Mind (ToM) -- often defined as the ability to reason about others' mental states -- has sparked significant scientific and public...
arxiv.org
March 6, 2025 at 6:05 PM
Turns out it isn't quite so straightforward: arxiv.org/abs/2502.21098
It's just my first impression. I asked it to act as a psychologist conducting an In-Depth-Interview (IDI) for a consumer good. It stayed in-character (great), but would get stuck in rabbit holes. Latency and EQ are great though.
March 2, 2025 at 3:48 PM
It's just my first impression. I asked it to act as a psychologist conducting an In-Depth-Interview (IDI) for a consumer good. It stayed in-character (great), but would get stuck in rabbit holes. Latency and EQ are great though.
It's sometimes trying a little too hard to be human, and I wonder how it'd do on instruction following.
March 2, 2025 at 12:38 PM
It's sometimes trying a little too hard to be human, and I wonder how it'd do on instruction following.
What's the unit on the axis? # vibes? vibes/token?
February 27, 2025 at 9:18 PM
What's the unit on the axis? # vibes? vibes/token?
For certain writing tasks, it is much better. Counting by the token doesn't do justice to the net tokens that actually end up in your final deliverable.
February 27, 2025 at 9:16 PM
For certain writing tasks, it is much better. Counting by the token doesn't do justice to the net tokens that actually end up in your final deliverable.
So if unemployment↑ and inflation↑ what will Fed do?
February 20, 2025 at 10:06 PM
So if unemployment↑ and inflation↑ what will Fed do?
Training for a new domain, with verifiable answers / solutions
February 11, 2025 at 3:09 PM
Training for a new domain, with verifiable answers / solutions
@weightsbiases.bsky.social about to rename?
February 4, 2025 at 8:12 AM
@weightsbiases.bsky.social about to rename?
MarkText can help with cleaning up markdown, it's a good editor. But I just use Cursor, even on .md files in an Obsidian vault.
January 31, 2025 at 10:09 PM
MarkText can help with cleaning up markdown, it's a good editor. But I just use Cursor, even on .md files in an Obsidian vault.
Notion? It's plain markdown, you can link between the markdown notes, and you can share.
January 31, 2025 at 8:43 PM
Notion? It's plain markdown, you can link between the markdown notes, and you can share.
In consumer good context, 10.1287/mksc.2022.1371 show that generational share gap is mostly due to experience, and less due to intrinsic differences.
January 23, 2025 at 1:44 PM
In consumer good context, 10.1287/mksc.2022.1371 show that generational share gap is mostly due to experience, and less due to intrinsic differences.
Yes, like that. Maybe you could categorize the controls.
January 23, 2025 at 1:41 PM
Yes, like that. Maybe you could categorize the controls.