pcoz

4 ms·

1.
Show HN: Make a free 3.8B model as reliable as one 7× bigger at parsing data(pypi.org)
4 points by pcoz 5mo ago|1 comments
2.
Python library to solve certain P-hard problems in polynomial time(pypi.org)
1 points by pcoz 6mo ago|1 comments
3.
by pcoz 6mo ago
Some problems classified as #P-hard - where counting valid solutions is thought to require exponential time - can have a hidden structure that makes them solvable exactly, in polynomial time. This library finds that encoding automatically,
4.
by pcoz 8mo ago
Every act of biological generation has the same structure: something that carries information (an enzyme, a gene regulator, a sperm cell) engages with something receptive (a substrate, a strand of DNA, an egg) - and something new is born fr
5.
by pcoz 10mo ago
Most neural networks assume computation is instantaneous: an input arrives, a function runs, an output appears. Even with sequences, time is often modeled indirectly (via windowing, stacking, or recurrence), so the model still reacts rather
6.
by pcoz 10mo ago
Are we overusing neural networks for time-series analysis? Neural networks are incredible tools. But in many real-world time-series problems, they’re simply overkill. If your signal is driven by a handful of rhythms, trends, decays, or satu
7.
by pcoz 10mo ago
Built a small experiment around conditional compute by reasoning type. This repo (1) classifies prompts into 6 reasoning types (weighting/consensus/deduction/comparison/causal/lookup) via simple patterns, then (2) r
8.
Free Vibe Coding eBook(amazon.com)
3 points by pcoz 11mo ago|0 comments
9.
AI and the Art of Persuasion(fleetingswallow.com)
2 points by pcoz 1y ago|1 comments
10.
by pcoz 1y ago
AI and The Art of Persuasion *************** We built an AI script that models persuasion as navigation through two-dimensional space—advancing legal arguments while simultaneously managing emotional progression. The framework applies to an
11.
An alternative to knowledge graphs for storing loosely structured content(fleetingswallow.com)
7 points by pcoz 1y ago|2 comments
12.
by pcoz 1y ago
Traditional knowledge graphs fail when applied uniformly to mixed documentation types. Force a well-organized spec through the same extraction pipeline as chaotic Slack threads and you either over-process structured content or under-extract
13.
AI Script to Amalgamate Requirements(github.com)
1 points by pcoz 1y ago|0 comments
14.
by pcoz 1y ago
It's really a prompt engineering thing. Prompt engineering allows you to play off different areas of the LLM's training data against each other to produce a useful result by returning the intersection between the two sets (knowled
15.
New AI Native Language(github.com)
3 points by pcoz 1y ago|3 comments
16.
by pcoz 1y ago
I wrote this language to address the issue with letting AI into production environments, which is that the AI is unpredictable to a certain degree. The purpose of this language is to provide hard, defined constraints in which the AI's