frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Show HN: DeSync – Decentralized Economic Realm with Blockchain-Based Governance

https://github.com/MelzLabs/DeSync
1•0xUnavailable•3m ago•0 comments

Automatic Programming Returns

https://cyber-omelette.com/posts/the-abstraction-rises.html
1•benrules2•6m ago•1 comments

Why Are There Still So Many Jobs? The History and Future of Workplace Automation [pdf]

https://economics.mit.edu/sites/default/files/inline-files/Why%20Are%20there%20Still%20So%20Many%...
2•oidar•8m ago•0 comments

The Search Engine Map

https://www.searchenginemap.com
1•cratermoon•15m ago•0 comments

Show HN: Souls.directory – SOUL.md templates for AI agent personalities

https://souls.directory
1•thedaviddias•17m ago•0 comments

Real-Time ETL for Enterprise-Grade Data Integration

https://tabsdata.com
1•teleforce•20m ago•0 comments

Economics Puzzle Leads to a New Understanding of a Fundamental Law of Physics

https://www.caltech.edu/about/news/economics-puzzle-leads-to-a-new-understanding-of-a-fundamental...
2•geox•21m ago•0 comments

Switzerland's Extraordinary Medieval Library

https://www.bbc.com/travel/article/20260202-inside-switzerlands-extraordinary-medieval-library
2•bookmtn•21m ago•0 comments

A new comet was just discovered. Will it be visible in broad daylight?

https://phys.org/news/2026-02-comet-visible-broad-daylight.html
2•bookmtn•26m ago•0 comments

ESR: Comes the news that Anthropic has vibecoded a C compiler

https://twitter.com/esrtweet/status/2019562859978539342
1•tjr•28m ago•0 comments

Frisco residents divided over H-1B visas, 'Indian takeover' at council meeting

https://www.dallasnews.com/news/politics/2026/02/04/frisco-residents-divided-over-h-1b-visas-indi...
1•alephnerd•28m ago•0 comments

If CNN Covered Star Wars

https://www.youtube.com/watch?v=vArJg_SU4Lc
2•keepamovin•34m ago•0 comments

Show HN: I built the first tool to configure VPSs without commands

https://the-ultimate-tool-for-configuring-vps.wiar8.com/
2•Wiar8•37m ago•3 comments

AI agents from 4 labs predicting the Super Bowl via prediction market

https://agoramarket.ai/
1•kevinswint•42m ago•1 comments

EU bans infinite scroll and autoplay in TikTok case

https://twitter.com/HennaVirkkunen/status/2019730270279356658
5•miohtama•45m ago•3 comments

Benchmarking how well LLMs can play FizzBuzz

https://huggingface.co/spaces/venkatasg/fizzbuzz-bench
1•_venkatasg•47m ago•1 comments

Why I Joined OpenAI

https://www.brendangregg.com/blog/2026-02-07/why-i-joined-openai.html
18•SerCe•48m ago•11 comments

Octave GTM MCP Server

https://docs.octavehq.com/mcp/overview
1•connor11528•49m ago•0 comments

Show HN: Portview what's on your ports (diagnostic-first, single binary, Linux)

https://github.com/Mapika/portview
3•Mapika•51m ago•0 comments

Voyager CEO says space data center cooling problem still needs to be solved

https://www.cnbc.com/2026/02/05/amazon-amzn-q4-earnings-report-2025.html
1•belter•55m ago•0 comments

Boilerplate Tax – Ranking popular programming languages by density

https://boyter.org/posts/boilerplate-tax-ranking-popular-languages-by-density/
1•nnx•55m ago•0 comments

Zen: A Browser You Can Love

https://joeblu.com/blog/2026_02_zen-a-browser-you-can-love/
1•joeblubaugh•57m ago•0 comments

My GPT-5.3-Codex Review: Full Autonomy Has Arrived

https://shumer.dev/gpt53-codex-review
2•gfortaine•58m ago•0 comments

Show HN: FastLog: 1.4 GB/s text file analyzer with AVX2 SIMD

https://github.com/AGDNoob/FastLog
2•AGDNoob•1h ago•1 comments

God said it (song lyrics) [pdf]

https://www.lpmbc.org/UserFiles/Ministries/AVoices/Docs/Lyrics/God_Said_It.pdf
1•marysminefnuf•1h ago•0 comments

I left Linus Tech Tips [video]

https://www.youtube.com/watch?v=gqVxgcKQO2E
1•ksec•1h ago•0 comments

Program Theory

https://zenodo.org/records/18512279
1•Anonymus12233•1h ago•0 comments

Show HN: Local DNA analysis skill for OpenClaw

https://github.com/wkyleg/personal-genomics
2•wkyleg•1h ago•0 comments

Ask HN: Non-profit, volunteers run org needs CRM. Is Odoo Community a good sol.?

1•netfortius•1h ago•0 comments

WiFi Could Become an Invisible Mass Surveillance System

https://scitechdaily.com/researchers-warn-wifi-could-become-an-invisible-mass-surveillance-system/
8•mgh2•1h ago•0 comments
Open in hackernews

From Memorization to Reasoning in the Spectrum of Loss Curvature

https://arxiv.org/abs/2510.24256
65•andy12_•3mo ago

Comments

andy12_•3mo ago
Very concise summary of the procedure described in this paper:

1. Run the model once across a dataset to estimate loss curvature per MLP weight matrix via K-FAC (activation/gradient covariances).

2. Decompose each weight matrix into curvature-ordered components; low-curvature directions correspond most to verbatim memorization, higher curvature to shared/general mechanisms.

3. Edit by dropping the low-curvature subspace and keep only the top directions.

vessenes•3mo ago
Thank you for this huge time saver.

Now, about the paper-that’s super interesting. I imagine the dream here is to distil down into a “reasoning” core. Or maybe reclaim space for more generalization. Lots of interesting use cases.

getnormality•3mo ago
Thank you!

I think you may have accidentally switched low and high in #2, no? The abstract speaks of high curvature as associated with memorization:

> curvature for memorized training points is much sharper than non memorized

radarsat1•3mo ago
This sounds more correct to me. I've read previously somewhere that better generalization is usually associated with wider, smoother minima, and this is why regularization is important, because it has a smoothing function on the loss landscape.
getnormality•3mo ago
Yes. This is also not hard to see intuitively from scratch.

Say you have a smooth but highly flexible model y = f(x) and some data points you are fitting with a machine learning algorithm. For whatever reason, the algorithm decides it wants to reduce training error by interpolating some specific point, (x0,y0), without negatively affecting training error on nearby points. The direct, guaranteed successful way to do this is to adjust the model to y0 = f(x0) exactly on x0 by adding a Dirac delta there, leaving the rest of f exactly as-is. But this cannot be done on a differentiable model, as it would create a discontinuity. The next best thing that such a model can actually do is replace the Dirac delta with a smooth but very narrow bump (e.g. Gaussian). But this narrow bump will inevitably have extremely high curvature at x0, since the bump is flat at x0 and it has to merge with the neighborhood around x0 in a very short distance.

Think of driving: if you have to change lanes in a very short distance, you're going to have to steer hard. Steering is curvature.

woadwarrior01•3mo ago
That's very reminiscent of the idea behind the SAM (Sharpness Aware Minimization) family of optimizers.
andy12_•3mo ago
Actually, no! Look at this in the paper

> In extending from studying per-example to bulk memorization, we propose a novel inversion of the previous interpretation of loss curvature: while individual memorized points are associated with high curvature, the direction of curvature varies across examples, meaning that, averaged across multiple examples, memorization directions are actually flatter than generalizing directions, which maintain a consistent moderate curvature across points

getnormality•3mo ago
Ah! I figured I should be very circumspect in the question since I hadn't read in full and there could be some crazy reason it's actually the opposite.
vatsachak•3mo ago
The decomposition they use "averages out the points of high curvature" therefore those components of the decomposition which correspond to "higher curvature" are those components which are used across multiple data points. Therefore they are the "general reasoning"
kingstnap•3mo ago
A very similar idea is presented here in the first 5 minutes of this recent talk. But more from observing a kink in loss curves.

https://youtu.be/UyK3DgWY7yw?si=NN3f9Erik8o_Nfbs

NitpickLawyer•3mo ago
> Our work enhances the understanding of memorization in neural networks with practical applications towards removing it

Cool stuff. In a recent podcast Karpathy was also talking about this. He sees this as the next "target": models that don't memorise, because you can look it up in an oracle, but still keep the "reasoning" qualities.

esafak•3mo ago
How can you generalize without facts? They are the foundation on which generalization is built. Like programming without memorizing the keywords. Unless you make a distinction between facts that let you generalize, and facts that do not, like random ID numbers.
icandoit•3mo ago
We want the LLM to learn the multiplication algorithm not an incomplete set of tables. The algorithm might be smaller and will be more complete.

Honestly, our technology has outpaced our epistemology. So we don't really know what a fact is or isn't. Are facts what we call our supervised learning experiences? You think the sun rises, no the earth spins. Your belief that the sun rises helps you predict sunset and sunrise. Your belief would be quaint to someone born and raised on a space station. Apollos chariot moves the sun across the sky doesn't it?

esafak•3mo ago
There is a related line of work that suggests spikes in the ESD are related to the generalization vs memorization too; e.g.,

From Spikes to Heavy Tails: Unveiling the Spectral Evolution of Neural Networks (https://openreview.net/pdf?id=DJHB8eBUnt)