Hacker News: john_horton

Add, delete and move data points to create a particular regression line

john_horton — Mon, 08 Dec 2025 18:26:48 +0000

Article URL: https://line-fitter--johnhorton.replit.app/

Comments URL: https://news.ycombinator.com/item?id=46195819

Points: 1

# Comments: 1

The Coasean Singularity? Demand, Supply, and Market Design with AI Agents

john_horton — Sun, 16 Nov 2025 20:28:09 +0000

Article URL: https://www.nber.org/papers/w34468

Comments URL: https://news.ycombinator.com/item?id=45948121

Points: 2

# Comments: 0

Superrational Reasoning in the Prisoner's Dilemma with LLMs

john_horton — Sat, 01 Nov 2025 17:12:15 +0000

Article URL: https://www.expectedparrot.com/content/clajelli/superrationality-game-theory

Comments URL: https://news.ycombinator.com/item?id=45783363

Points: 2

# Comments: 0

A Poor Man's User Study with a Vision Model and E[P]

john_horton — Fri, 11 Jul 2025 00:58:26 +0000

Article URL: https://twitter.com/johnjhorton/status/1943473769219002766

Comments URL: https://news.ycombinator.com/item?id=44527391

Points: 1

# Comments: 0

A little bit of human-provided structure gives better LLM answers

john_horton — Sat, 03 May 2025 22:23:18 +0000

Article URL: https://blog.expectedparrot.com/p/whit-diffie-erasure

Comments URL: https://news.ycombinator.com/item?id=43882857

Points: 1

# Comments: 0

Using vision capable LLMs to transcribe a handwritten letter

john_horton — Mon, 17 Feb 2025 21:09:29 +0000

Article URL: https://www.expectedparrot.com/content/johnjhorton/grant-letter

Comments URL: https://news.ycombinator.com/item?id=43083309

Points: 2

# Comments: 0

New comment by john_horton in "Ask HN: What is the best method for turning a scanned book as a PDF into text?"

john_horton — Mon, 17 Feb 2025 19:09:07 +0000

This might be of interest: https://www.expectedparrot.com/content/johnjhorton/grant-let... It uses a collection of models to extract the text from a handwritten letter by US Grant. Probably overkill for something nicely printed.

Little's Law

john_horton — Wed, 02 Oct 2024 23:00:40 +0000

Article URL: https://en.wikipedia.org/wiki/Little%27s_law

Comments URL: https://news.ycombinator.com/item?id=41725707

Points: 3

# Comments: 0

LaTeX writing checker in Python using LLMs

john_horton — Thu, 15 Aug 2024 15:38:00 +0000

Article URL: https://www.expectedparrot.com/content/e8012216-2f36-42c2-9008-9ee040ffcf36?from_page=my_content

Comments URL: https://news.ycombinator.com/item?id=41257302

Points: 2

# Comments: 0

New comment by john_horton in "Python Library for Structured Data Extraction via LLM"

john_horton — Wed, 14 Aug 2024 15:11:45 +0000

Hey thanks for noticing - here's the MIT licensed library it's based on: https://github.com/expectedparrot/edsl

New comment by john_horton in "Launch HN: Trellis (YC W24) – AI-powered workflows for unstructured data"

john_horton — Wed, 14 Aug 2024 11:15:11 +0000

thanks! B/c it got some positive reaction here, I did a little thread on how you can turn this flow into an API: https://x.com/johnjhorton/status/1823672992624242895

New comment by john_horton in "Launch HN: Trellis (YC W24) – AI-powered workflows for unstructured data"

john_horton — Tue, 13 Aug 2024 23:57:51 +0000

Very cool - I've been working on an open source python package that lets you do some similar things (https://github.com/expectedparrot/edsl).

Here's an example of the Enron email demo using the edsl syntax/package & a few different LLMs: https://www.expectedparrot.com/content/6607caa1-efc5-439f-85...

New comment by john_horton in "Python package for administering surveys to LLMs"

john_horton — Thu, 18 Apr 2024 14:26:29 +0000

Link to the package: https://github.com/expectedparrot/edsl

Python package for administering surveys to LLMs

john_horton — Thu, 18 Apr 2024 14:23:39 +0000

Article URL: https://twitter.com/johnjhorton/status/1780937297967423707

Comments URL: https://news.ycombinator.com/item?id=40076613

Points: 1

# Comments: 1

Algorithmic Writing Assistance on Jobseekers’ Resumes Increases Hires

john_horton — Sun, 29 Jan 2023 16:06:25 +0000

Article URL: https://twitter.com/emmavaninwegen/status/1618714903371722752

Comments URL: https://news.ycombinator.com/item?id=34569313

Points: 2

# Comments: 0

New comment by john_horton in "Large language models as simulated economic agents (2022) [pdf]"

john_horton — Sun, 15 Jan 2023 14:56:10 +0000

Got it - so it is the "performativity critique" - the idea that the LLM "knows" economic theories and responds in accordance with those theories. I don't think that's very likely because a) econ writing is presumably a tiny, tiny fraction of the corpus and (b) it would imply an amazing degree of transfer learning e.g., it would know to apply "status quo bias" (because it ready the papers) to new scenarios. But as the paper makes clear, you can't use it to "confirm" theories but rather use it like economists use other models - to explore behavior and generate testable predictions cheaply that you can go test with actual humans in realistic scenarios. The last experiment in the paper is from an experiment in a working paper of mine. There's no way the LLM knows this result, but if I had reverse the temporal order (create the scenario w/ the LLM, then run the experiment), it could have guided what to look at. That's likely what's scientifically useful. Anyway, thanks for engaging.

New comment by john_horton in "Large language models as simulated economic agents (2022) [pdf]"

john_horton — Sun, 15 Jan 2023 03:30:46 +0000

I'm sorry I don't follow - is your claim, that, say, an AI agent exhibiting status quo bias in responding to decision scenarios (e.g., a preference for options posed as the status quo relative to a neutral framing - Figure 3) that the reason this happens, empirically, is because the LLM has been trained on text describing status quo bias? E.g., like if an apple fell to the ground in an game, it was because the physics engine had been programmed w/ laws of gravity?

New comment by john_horton in "Large language models as simulated economic agents (2022) [pdf]"

john_horton — Sun, 15 Jan 2023 01:58:27 +0000

Are you referring to this: "What kinds of experiments are likely to work well? Given current capabilities, games with complex instructions are not presently likely to work well, but with more advanced LLMs on the horizon, this is likely to change. I should also note that research questions like what is “the effect of x on y” are likely to work much better than questions like “what is the level of x?.” Consider that in my Kahneman et al. (1986) example, I can create AI “socialists” who are not too keen on the price system generally. If I polled them about who they want for president, there is no reason to think it would generalize to the population at large. But if my research question was “what is the effect of the size of the price increase on moral judgments” I might get be able to make progress. That being said, it might be possible to create agents with the correct “weights” to get not just qualitative results but also quantitatively accurate results. I did not try, but one could imagine choosing population shares for the Charness and Rabin (2002) “types” to match moments with reality, then using that population for other scenarios." --- To clarify, this about what research questions are likely to work well here, not what questions posed to LLMs will work well.

New comment by john_horton in "Large language models as simulated economic agents (2022) [pdf]"

john_horton — Sat, 14 Jan 2023 23:49:20 +0000

It is so implausible that the training process that creates LLMs might learn features of human behavior that could then be uncovered via experimentation? I showed, empirically, that one can replicate several findings in behavioral economics with AI agents. Perhaps the model "knows" how to behave from these papers, but I think the more plausible interpretation is that it learned about human preferences (against price gouging, status quo bias, & so on) from its training. As such, it seems quite likely that there are other latent behaviors captured by LLMs and yet to be discovered.

New comment by john_horton in "Large language models as simulated economic agents (2022) [pdf]"

john_horton — Sat, 14 Jan 2023 20:33:22 +0000

that's a great suggestion - thanks!