birb_cromble
No bio...
User ID: 3236
birb_cromble
8hr ago
That's interesting. According to family, I was non-verbal until I was about three and a half, then immediately started speaking in complete, grammatically-correct sentences.
birb_cromble
11hr ago
For a while, Gemini had patched this, but if you said you were driving a truck or changed the distance, it would revert right back to the broken behavior.
birb_cromble
11hr ago
Any opinions on the SEC's proposed reporting requirement changes? The section on NAFs, specifically, seems bad for retail investors. It looks like it's reducing reporting requirements for companies that qualify.
birb_cromble
15hr ago
Are they lying? Was the kernel made up?
Cases like this, and the erdos problems, are exactly where LLMs shine. Problems with clear and unambiguous reward functions that are difficult to hack are perfect use cases. In the Alibaba case, they likely have an extensive set of characterization tests that guarantee consistent behavior. An LLM with a good harness can pound its head against those tests forever while simultaneously measuring the performance as a success metric. It will never get tired and it won't get sick of doing that kind of work.
There's definitely value there, but I don't know how much value. The combination of technical depth and strong guardrails make for a very schizophrenic kind of difficulty. Doing that kind of work is traditionally either the domain of a plucky junior with too much energy, or an insane wizard who claimed a broom closet as his office.
When we've experimented with that kind of optimization work at my employer, it tends to be very expensive, since most of the results come from the absolute tirelessness of the agent. In comparison, how much are you paying your junior? How much are you paying your wizard, and what is he doing if he's not doing that task? Security scans are a similar thing. Line audits aren't hard, but they're hella time consuming. As model costs rise (and they are rising per task completed when you compare any single vendor over time), it might legitimately be cheaper to throw interns at the problem than LLMs.
At least on the software side, I think there's a reasonable chance that what we're seeing is a temporary pop due to a lot of highly verifiable technical debt deadwood finally getting burned out, and that might not be a constant source of demand.
On the war side, I wish I knew more. The sensitive nature of the topic means that all parties are incentivized to obfuscate and dissemble as much as possible. It might legitimately be an ideal case. LLMs do well when you can accept 95% accuracy, and in something like intelligence analysis, 95% accuracy probably has the spooks all but shitting their pants.
birb_cromble
16hr ago
I don't know if there was ever really a transition. It's been something that I've had to consider my entire life.
For anyone out there who has a drinking problem, seek help. If you have kids, do it sooner, rather than later.
birb_cromble
1d ago
Can you elaborate a little on the real estate collapse that you mention? You bring it up a few times like it's common knowledge, but I've never read anything about it.
birb_cromble
1d ago
They're firmly on my "investigate" list.
birb_cromble
1d ago
England declared war on France and nobody was happy about it.
I've expressed here in the past that I'm not seeing the huge gains from LLM based coding that a lot of other parties are seeing. A new research paper suggests that maybe I'm not just prompting it wrong.
We present a systematic study evaluating how well agents handle structural constraints in multi-file backend generation. By fixing a unified API contract across 80 greenfield generation tasks and 20 feature-implementation tasks spanning eight web frameworks, we isolate the effect of structural complexity using a dual evaluation with end-to-end behavioral tests and static verifiers. Our findings reveal a phenomenon of constraint decay: as structural requirements accumulate, agent performance exhibits a substantial decline. Capable configurations lose 30 points on average in assertion pass rates from baseline to fully specified tasks, while some weaker configurations approach zero.
Specifically, the paper calls out convention based frameworks and ORMs as particularly problematic, and I'm using both. It feels good to see some evidence that I'm not crazy and incompetent.
birb_cromble
1d ago
I was pretty autistic on that test - in the same range as @ToaKraka, who seems to be our gold standard here on TheMotte.
I was also diagnosed as an adult when seeking treatment for PTSD. Up until that point, I had no idea. Growing up, there were a lot of comments about how "that boy ain't right" and how I needed to "act normal". The idea that it might be something diagnosable or treatable didn't exist in those communities. Instead of an IEP, I got my ass beat until I could fake it well enough to get by.
Now that I know, it doesn't really change much. Mostly it just informed the PTSD treatment. Well, it informs treatment and gives my partner a new way to playfully make jokes about my behavior at times.
TK: Please don't take the gold standard comparison as an insult. You're one of my favorite people on this forum. The weird shit you dig up and post brightens my day whenever I see it.
birb_cromble
1d ago
Now that you mention it, when was the last time you even saw a minimum wage job? Even the convenience stores and McDonald's around me are offering well over the state minimums. It seems like it's one of those things that exists on paper but doesn't really come up anymore.
birb_cromble
1d ago
Anecdotally, I'm seeing a lot more of that around me. Punk is making a raging comeback.
It might be worth moving this over into the finance thread, but I am at least partially putting my money where my mouth is.
I'm of the opinion that the current LLM hype is starting to hit the second knee of the S-curve, both financially and technically.
Technically, exponential growth leads to exponential friction, and it looks to me like the real-world improvements in model capabilities are slowing down between generations. Anecdotally, it feels like the models are increasingly fungible and most of the ostensible improvements have come from harnesses, which are regular old software engineering. I think there's something there, but I think LLM tech represents a local maximum. I'm eagerly watching whatever Yann Lecun is cooking up at AMIL, because the general concept of a world model seems to map better to what we think of as "intelligence". His paper on energy models, specifically, is fascinating.
Financially, I think a lot about the "during a gold rush, sell shovels" aphorism. I also think about Buffett and Munger's rules of investing. Meta and Oracle are buying shovels, but using them to dig their own graves, so far as I can tell. I don't think Anthropic and OpenAI are ever going to be able to support their valuations, and per their S-1, xAI has already pretty much given up. If Google dies, it'll be for reasons other than AI spending. Nvidia has a good product and a good moat for now, but various specialized competitors are nipping at their heels while Chinese cards may develop into direct competitors.
In other words, I think the tech is going to continue developing, but I think a lot of the current players are in for a rude reminder of market on market fundamentals by 2H 2027 or so. I know you've bemoaned "financialization" in the past, but at the end or the day the economy is just people buying and selling things, and fuck me if it doesn't seem like some of these companies are trying to act like that's not true.
Where does that leave me? I'm moving down the stack. If the tech is going somewhere, it has to run on real things and interact with the real world eventually. Companies doing physical things are riskier to start than pure software, but they're less likely to get disrupted once they establish themselves. I've largely stopped investing in funds that hold significant amounts of meta, oracle, Tesla (because I think they may absorb SpaceX), and even Nvidia. On the other hand, I'm expanding my positions in funds that hold TSMC, ASML, and Lam Research. I am watching Cerebras, but I won't invest until I better understand how they're using software to get around defects on their enormous chips.
Don't flame out because they haven't followed what you consider the optimal path. Everybody has to start somewhere.
I'm not a mod, but please don't lead with accusations of bad faith. It's clear you have a lot of experience in this area, and I fear you're assuming that the layman's level of expertise is much higher than it is.
birb_cromble
2d ago
past like 5+ years
2022 was an apocalyptically bad year for bonds. It was the worst year in history. The runner up was 1803. How does your four year look?
Beyond that, are you looking at share appreciation or total yield?
birb_cromble
2d ago
You can pry my minus sign dash from my cold, dead hands.
birb_cromble
2d ago
It could still be, but I'm too old to keep track of that shit anymore.
birb_cromble
2d ago
People like MLID have been floating rumors about Intel killing their consumer GPUs since about ten seconds after ARC came out. At this point I take rumors of cancellation as evidence of an imminent release.
birb_cromble
2d ago
For tax loss harvesting specifically there's a limit, but I'm going to go out on a limb and assume that somebody with Trump's net worth has access to tax avoidance schemes that don't even have names yet.
birb_cromble
2d ago
How does the price of SLV and the NAV of SLV correlate to the spot price of physical silver? Reading the prospectus, it seems like it is actually holding a fair amount of physical silver in trust, but I'm out traveling for the weekend and I can't really dig.
birb_cromble
2d ago
I think we might be talking past each other. I was talking about treasury inflation protected securities.
Charitably, it could be tax loss harvesting. At high net worth and income strategies like that can get pretty intense.
birb_cromble
2d ago
I wish TIPS weren't taxed the way they are. It seems almost perfectly self-defeating
birb_cromble
2d ago
This is going to sound insane, but learn to write music in standard notation. I'm a lot of ways, it's a very simplified programming language.
An old anthology of Lovecraftian Horror. It's interesting to see how views of both the supernatural and technology have changed since its publishing.
More options
Context Copy link