xianshou

Ironically, "load-bearing" appears in the main text of the article that says never to use a single word the LLM suggests. Womp womp.

the future is here and one should be thankful for its slightly uneven distribution. otherwise we would hardly have anything left about which to develop strong opinions!

Really missed the opportunity for "Airemin".

A lovely example of a study that is both obviously true and misses the point.

Music with lyrics directly interferes with any task that has a verbal component, and the worse you are at multitasking, the worse the interference. Despite being terrible at multitasking, I still listen to music with lyrics. Why? Principally because the alternative, hearing all the conversations in my immediate vicinity, is usually both more distracting and less pleasant. But there are also auxiliary benefits, such as an increase in "work stamina" and a passive signal to coworkers to interrupt only if it's important.

Now, I could listen to lo-fi all day, or three-hour soundtracks on Youtube, and sometimes do, but it gets boring pretty fast!

Anyway: obviously true, still worth it because the alternative is worse.

(By the way, other mitigating strategies: listening to music in a language you don't understand, or listening to lyrics so familiar you can screen them out. My top Spotify songs all get played several hundred times a year.)

Even as someone extremely firmly on the other side of the AI debate, I must appreciate the craft.

Now, to give Claude the steganogravy skill...

From the file: "Answer is always line 1. Reasoning comes after, never before."

LLMs are autoregressive (filling in the completion of what came before), so you'd better have thinking mode on or the "reasoning" is pure confirmation bias seeded by the answer that gets locked in via the first output tokens.

I appreciate not having to read this guy again.

Great work! Why no benchmarks though?

Nice! 5 bucks says you can swap this in for your average software kanban and it does a better job.

Safer than clawdbot/moltbot, I'll bet.