frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

I built a Million Dollar Homepage for text, and it's chaotic

https://www.themillionlines.com
1•lolzenom•1m ago•1 comments

Show HN: Kerns – A Continuous Research Workspace

https://www.kerns.ai/
2•kanodiaayush•2m ago•0 comments

Free Culture book: programming essentials with Scheme

https://www.draketo.de/software/programming-scheme
1•ArneBab•4m ago•1 comments

LLM-powered What If text gen for fun

1•techbuilder4242•5m ago•0 comments

Fidji Simo: ChatGPT Health and what AI can do for a broken system

https://fidjisimo.substack.com/p/chatgpt-health
1•nadis•5m ago•0 comments

Minneapolis Mayor Blasts Kristi Noem's BS ICE Shoots Kills Woman in Minnesota

https://www.thedailypoliticususa.com/p/minneapolis-mayor-blasts-kristi-noems
9•Jacquie11•6m ago•4 comments

There Is a Sickness Eating Away at American Democracy

https://www.nytimes.com/2026/01/06/opinion/trump-jan-6-jefferson-davis.html
3•whack•8m ago•0 comments

How being on LinkedIn feels like now

https://plonkedin.vercel.app/
2•falloutx•8m ago•0 comments

Cursor: Dynamic Context Discovery

https://cursor.com/blog/dynamic-context-discovery
2•leerob•9m ago•0 comments

The Films of 2025:Q4

https://scottsumner.substack.com/p/the-films-of-2025q4
1•paulpauper•10m ago•0 comments

Teenygrad

https://github.com/tinygrad/teenygrad
1•jxmorris12•10m ago•0 comments

Persuasion of Humans Is the Bottleneck

https://erikschiskin.substack.com/p/persuasion-of-humans-is-the-bottleneck
1•paulpauper•10m ago•0 comments

CKSyncEngine Questions and Answers

https://christianselig.com/2026/01/cksyncengine/
1•chmaynard•12m ago•0 comments

The Philosophy of Solvej Balle

https://endsdontjustifythemeans.com/p/the-philosophy-of-solvej-balle
1•paulpauper•12m ago•0 comments

CheckMyLLM – A real-time "status board" for LLM reliability

https://checkmyllm.com/
2•valpine8•12m ago•2 comments

Longbeard: Catholic Social Teaching and AI

https://www.longbeard.com/blog/catholic-social-teaching-and-ai
1•admaiorem•12m ago•1 comments

Recipe for a great startup dev team

https://renaissance.kelsus.com/p/recipe-for-a-great-startup-dev-team
1•nadis•14m ago•0 comments

Show HN: AI Swarm v3 – Self-host your own headless AI agents

https://ai-swarm.dev
1•plurb-unus•16m ago•0 comments

HP Reveals Keyboard Computer with Ryzen AI Chip

https://www.hp.com/us-en/desktops/business/eliteboard.html
2•tonymet•17m ago•0 comments

Antiwar AI

https://nikonole.com/antiwarai
2•throwoutway•19m ago•0 comments

Democratizing 3D for Everyone

https://www.youtube.com/watch?v=oGOkx7cuwvo
2•AmanPorwal•20m ago•1 comments

Show HN: SpaceXYZ

https://angular-audio.com/space-xyz
1•jsmithoner•24m ago•0 comments

How AI Is Learning to Think in Secret

https://nickandresen.substack.com/p/how-ai-is-learning-to-think-in-secret
1•NickAndresen•27m ago•2 comments

Applied Mathematics Notebooks Repository:Collection of Google Colab Notebooks

https://github.com/GirolamoOddo/AppliedMath_Notebooks
1•joebig•27m ago•0 comments

What 'I'm an American living in Australia' trend reveals about both countries

https://www.abc.net.au/news/2026-01-08/what-im-an-american-in-australia-trend-reveals-about-usa-a...
1•defrost•29m ago•0 comments

Trump Crypto Venture World Liberty Applies for Bank Charter

https://www.bloomberg.com/news/articles/2026-01-07/trump-crypto-venture-world-liberty-applies-for...
2•geox•29m ago•0 comments

AI chip frenzy to wallop DRAM prices with 70% hike

https://www.theregister.com/2026/01/06/memory_firm_profits_up_as/
3•ironyman•29m ago•0 comments

Objective-See: Lulu

https://objective-see.org/products/lulu.html
2•throwoutway•37m ago•0 comments

Some Gleam Projects

https://dusty.phillips.codes/2026/01/05/some-gleam-projects/
2•todsacerdoti•41m ago•0 comments

Trying to Launch into 2026

https://www.bennadel.com/blog/4858-trying-to-launch-into-2026.htm
1•sea-gold•43m ago•0 comments
Open in hackernews

Mathematical framework 4 cntrling LLM behavior via information-geometric Torsion

https://github.com/merchantmoh-debug/torsion-control-network
1•MohskiBroskiAI•1d ago

Comments

MohskiBroskiAI•1d ago
Overview

Torsion Control Network (TCN) is a provably stable framework for controlling large language model behavior using:

Information geometry: Treat LLM outputs as manifold trajectories

Active inference: Free energy minimization for alignment

Torsion tensors: Mathematical operators for bending response distributions

Lyapunov stability: Formal guarantees of convergence to desired behavior

The Problem

Current LLM alignment methods:

RLHF: Unstable, requires massive compute, mode collapse

Constitutional AI: Brittle rules, easy to jailbreak

Prompt engineering: Ad-hoc, no guarantees, prompt injection vulnerabilities

Fine-tuning: Catastrophic forgetting, distribution shift

The Solution TCN provides mathematical control via:

Geodesic flow: Model LLM generation as trajectory on probability manifold

Torsion injection: Apply curvature to steer toward desired outputs

Free energy minimization: Stability via thermodynamic principles

Provable convergence: Lyapunov function guarantees alignment

Result: 95% alignment success with 1000x less compute than RLHF.

yorwba•1d ago
Did you forget to commit and push the files listed under "repository structure"?
MohskiBroskiAI•1d ago
Ah. Good eye m8. I let my comet agent do all the work. If there's any adminstrative errors in the setup blame him not me.

I'm a lazy genius. I like doing 20 repos in 30 minutes but God Forbid I actually learn how to do it myself >.>

yorwba•5h ago
Your agent is a lazy idiot. Didn't do the work but pretended everything was fine. It wasn't fine and it still isn't. You'll have to become paranoid. Check every single file. Every single line. Don't trust, verify. Otherwise you'll look like a fool when you try to show off the work the agent was supposed to have done, but there's nothing worth looking at.
MohskiBroskiAI•23h ago
Alright fixing it right now.Check again in 2 hours max.
MohskiBroskiAI•22h ago
Fixed.