Hacker News: trash_cat

New comment by trash_cat in "LLMs as a Cognitive Virus"

trash_cat — Sun, 06 Sep 2026 12:06:02 +0000

I think most of the comments are talking about the some "virus" but the actual valuable part of the paper is about making choices when interacting with LLMs in such a way that abilities are reversable. That you are still able to do those things without LLMs. And that society and institutions help to preserve this ability by encouraging ways to do things with and without LLMs.

New comment by trash_cat in "Understanding ChatGPT Work"

trash_cat — Mon, 31 Aug 2026 12:45:16 +0000

Why not use Codex for that? What is the differnce between work and codex for a very specific task that involves terminals?

New comment by trash_cat in "I accidentally turned LLM memory into program analysis"

trash_cat — Sun, 30 Aug 2026 16:40:57 +0000

                       agent memory
                             |
              +--------------+--------------+
              |                             |
       deductive state               episodic memory
              |                             |
       facts / rules / time          fuzzy context
       provenance                    semantic retrieval
       retractions                   source text

Well this turns out, just like human memory functions. (Semantic vs episodic memory)

New comment by trash_cat in "AMD acquires Taalas to boost inference performance by etching models in silicon"

trash_cat — Fri, 07 Aug 2026 09:26:20 +0000

People are not talking enough how huge this is for robotics and IoT. Current robotics arhitectures are limited by tok/sec. How cares if its not a Fable model?

This move undercuts NVIDIA directly.

New comment by trash_cat in "Pareto Front"

trash_cat — Thu, 06 Aug 2026 12:40:46 +0000

I think the one from hugging face is much clearer, albeit its an arena metric and a bit tricky to find the pareto view. Look for the the top Navigation Bar (Agent Chat Code Image Video). Chat -> (dropdown) Text -> (Side panel) View as Pareto.

https://huggingface.co/spaces/lmarena-ai/arena-leaderboard

New comment by trash_cat in "Situational Awareness and the Impending Stock Market Volatility"

trash_cat — Mon, 03 Aug 2026 10:48:57 +0000

"The forces that destroyed SA were also what generated its 4×+ return"...yes, that is what levrage means? And it goes both ways.

The interesting part what this article states: SA was essentialy a thematic ETF without any hedging to buffer downside, and got margin called.

New comment by trash_cat in "Running Kimi K3 on MI355X at Better Performance per Dollar Than B300"

trash_cat — Sun, 02 Aug 2026 15:57:39 +0000

Have $300k in disposible income?

New comment by trash_cat in "Show HN: I was tired of opening 2 tabs for every HN link, so I made a userscript"

trash_cat — Wed, 29 Jul 2026 07:56:51 +0000

So I have to Shift + Alt + n, the search for the HN blog / article in the other tab? If you have a lot of tabs this isn't feasable.

New comment by trash_cat in "Show HN: I was tired of opening 2 tabs for every HN link, so I made a userscript"

trash_cat — Wed, 29 Jul 2026 07:54:58 +0000

Wouldn't it make sense to have it an extension? Just spontainously (without reading any instructions) I don't understand how to use this or what Userscripts Tamper or Violentmonkey is.

New comment by trash_cat in "Our position on open-weights models"

trash_cat — Tue, 28 Jul 2026 06:59:08 +0000

How about;

1. Make anti-distillation clauses illegal to strenghten western opensource eco-system.

2. Make cyber-defensive models widely available (can detect but won't operationalize vulnerabilities). Otherwise make Fable awailable for everyone. Keeping the cybersecurity in assymetry by withholding cabailities just causes more instability and increases the incentives for powerfull close sourced models. This is a loss for everyone.

New comment by trash_cat in "You only need the frontier model for one single edit"

trash_cat — Tue, 21 Jul 2026 15:41:38 +0000

When a frontier makes a succesfull edit based on the plan that it made, it leaves an procedural trace in turn biases the NEXT model, low cost model, straight into procedural action. The cheaper model doesn't need to reread everything again because it has enough information from the frontier model to complete the task. A simple "plan" of what needs to be done does not carry this information.

New comment by trash_cat in "You only need the frontier model for one single edit"

trash_cat — Tue, 21 Jul 2026 15:32:17 +0000

Yes this is correct. It essentially biases the cheaper model into procedural action grounded by the frontier model. To put it mode generally, it injects the cheap model context with more useful information, where it might not need to read all the files again.

New comment by trash_cat in "Nvidia RTX Spark"

trash_cat — Wed, 03 Jun 2026 13:45:47 +0000

There is more to it than typing "llama serve".

New comment by trash_cat in "I think Anthropic and OpenAI have found product-market fit"

trash_cat — Thu, 28 May 2026 17:01:19 +0000

Discovering that Enterprise customers tolerate higher prices compared to retails consumers is discovering demand elasticities not PMF. Am I missig something?

New comment by trash_cat in "Turning a MacBook into a touchscreen with $1 of hardware (2018)"

trash_cat — Tue, 31 Mar 2026 02:13:26 +0000

Nobody forces you to use touchscreen exclusively?

New comment by trash_cat in "LinkedIn uses 2.4 GB RAM across two tabs"

trash_cat — Sun, 29 Mar 2026 17:05:44 +0000

Sales people using it a lot to scout prospects and understand a person's seniority in an organisation, to target better and prepare a strategy to pitch higher up the chain.

New comment by trash_cat in "CXMT has been offering DDR4 chips at about half the prevailing market rate"

trash_cat — Sun, 22 Feb 2026 08:50:06 +0000

Geopolitics and industrial policty aside, I think it's important to check how stable and reliable these chips are. I wouldn't count on them being on par with "western" ones. Correct me if I am wrong here.

New comment by trash_cat in "AI doesn’t reduce work, it intensifies it"

trash_cat — Tue, 10 Feb 2026 11:15:07 +0000

My two cents that this is part of the learning curve. With collective experience this type of work will be more understood, shared and explored. It is intense in the beginning because we are still discovering how to work with it. I think the other part being that this is a non-deterministic tool which does increase some cognitive load.

New comment by trash_cat in "Claude Opus 4.6"

trash_cat — Thu, 05 Feb 2026 18:15:29 +0000

I think the naming schemes are quite arbitrary at this point. Going to 5 would come with massive expectations that wouldn't meet reality.

New comment by trash_cat in "Iran Protest Death Toll Could Top 30k, According to Local Health Officials"

trash_cat — Sun, 25 Jan 2026 15:07:36 +0000

BitChat comes to mind.