Tootfinder

How to tell a vibe coder of lying when they say they check their code.
People who will admit to using LLMs to write code will usually claim that they "carefully check" the output since we all know that LLM code has a lot of errors in it. This is insufficient to address several problems that LLMs cause, including labor issues, digital commons stress/pollution, license violation, and environmental issues, but at least it's they are checking their code carefully we shouldn't assume that it's any worse quality-wise than human-authored code, right?
Well, from principles alone we can expect it to be worse, since checking code the AI wrote is a much more boring task than writing code yourself, so anyone who has ever studied human-computer interaction even a little bit can predict people will quickly slack off, stating to trust the AI way too much, because it's less work. I'm a different domain, the journalist who published an entire "summer reading list" full of nonexistent titles is a great example of this. I'm sure he also intended to carefully check the AI output, but then got lazy. Clearly he did not have a good grasp of the likely failure modes of the tool he was using.
But for vibe coders, there's one easy tell we can look for, at least in some cases: coding in Python without type hints. To be clear, this doesn't apply to novice coders, who might not be aware that type hints are an option. But any serious Python software engineer, whether they used type hints before or not, would know that they're an option. And if you know they're an option, you also know they're an excellent tool for catching code defects, with a very low effort:reward ratio, especially if we assume an LLM generates them. Of the cases where adding types requires any thought at all, 95% of them offer chances to improve your code design and make it more robust. Knowing about but not using type hints in Python is a great sign that you don't care very much about code quality. That's totally fine in many cases: I've got a few demos or jam games in Python with no type hints, and it's okay that they're buggy. I was never going to debug them to a polished level anyways. But if we're talking about a vibe coder who claims that they're taking extra care to check for the (frequent) LLM-induced errors, that's not the situation.
Note that this shouldn't be read as an endorsement of vibe coding for demos or other rough-is-acceptable code: the other ethical issues I skipped past at the start still make it unethical to use in all but a few cases (for example, I have my students use it for a single assignment so they can see for themselves how it's not all it's cracked up to be, and even then they have an option to observe a pre-recorded prompt session instead).

A quick reminder: The Standard Ticket sale for Berlin Buzzwords 2025 ends on Monday, June 9th!
If you're planning to attend in person from June 15-17, make sure to purchase your ticket before the deadline. Last Minute Tickets will be available afterwards at a higher rate.
Get ready for the conference – buy your ticket now: h…

Tickets
Join us for two days of exciting talks, workshops and discussions, and of course our barcamp on Sunday! Berlin Buzzwords will take place as a hybrid conference online and in our venue Kulturbrauerei in Berlin.

🦜 Gossip, redefined: Research shows the surprising power of positive talk in teams
#work

Gossip, redefined: Research shows the surprising power of positive talk in teams
Mean Girls and pop culture has taught us to fear gossip as toxic and cruel, the whispered fuel of cliques and drama. But what if we've been getting gossip all wrong? New research from UNM's organizational behavior team suggests that some gossip might actually help teams thrive.

As I'm learning Dutch, I'm reminded that the idea that there are people who believe that the bible is to be taken literally. The idea that a several hundred year old translation of a collection of texts in multiple languages, that were themselves translated multiple times between languages, before the whole thing was translated to Latin, then being translated to English, could somehow perfectly reflect the original text... Yeah, it's only possible to believe that if you have no idea how languages work and have never learned another language.
Like, just from linguistic drift alone if the bible were written in King James English you're losing *so* much context. But Hebrew, Aramaic, and Greek translated to Latin, then to English, then to English again?
There are so many things that erg can't be translated, even as a beginner. Dutch and English are two of the closest languages that exist, they're both Germanic languages and they're the closest to each other (other than Friesian). You can't really be much closer, and yet, there are so many things you can't mutually represent. Hebrew and Latin, Aramaic and Latin, Latin and English, Greek and English, these aren't even the same families at all... They're extremely distant. There's absolutely no way to represent concepts from one to another without another book's worth of explanation.
And that ignores all the cultural context, which is mostly lost and a library and decade of education to get the stuff that we *do* know.
Only monolingual Americans could come up with an idea so incredibly asinine.

Locating Risk: Task Designers and the Challenge of Risk Disclosure in RAI Content Work
Alice Qian Zhang, Ryland Shaw, Laura Dabbish, Jina Suh, Hong Shen
https://arxiv.org/abs/2505.24246

Locating Risk: Task Designers and the Challenge of Risk Disclosure in RAI Content Work
As AI systems are increasingly tested and deployed in open-ended and high-stakes domains, crowd workers are often tasked with responsible AI (RAI) content work. These tasks include labeling violent content, moderating disturbing text, or simulating harmful behavior for red teaming exercises to shape AI system behaviors. While prior efforts have highlighted the risks to worker well-being associated with RAI content work, far less attention has been paid to how these risks are communicated to wor…

How popular media gets love wrong
Okay, my attempt at (hopefully widely-applicable) advice about relationships based on my mental "engineering" model and how it differs from the popular "fire" and "appeal" models:
1. If you're looking for a partner, don't focus too much on external qualities, but instead ask: "Do they respect me?" "Are they interested in active consent in all aspects of our relationship?" "Are they willing to commit a little now, and open to respectfully negotiating deeper commitment?" "Are they trustworthy, and willing to trust me?" Finding your partner attractive can come *from* trusting/appreciating/respecting them, rather than vice versa.
2. If you're looking for a partner, don't wait for infatuation to start before you try building a relationship. Don't wait to "fall in love;" if you "fall" into love you could just as easily "fall" out, but if you build up love, it won't be so easy to destroy. If you're feeling lonely and want a relationship, pick someone who seems interesting and receptive in your social circles and ask if they'd like to do something with you (doesn't have to be a date at first). *Pursue active consent* at each stage (if they're not interested; ask someone else, this will be easier if you're not already infatuated). If they're judging you by the standards in point 1, this is doubly important.
3. When building a relationship, try to synchronize your levels of commitment & trust even as you're trying to deepen them, or at least try to be honest and accepting when they need to be out-of-step. Say things and do things that show your partner the things (like trust, commitment, affection, etc.) that are important in your relationship, and ask them to do the same (or ideally you don't have to ask if they're conscious of this too). Do these things not as a chore or a transaction when your partner does them, but because they're the work of building the relationship that you value for its own sake (and because you value your partner for themselves too).
4. When facing big external challenges to your commitment to a relationship, like a move, ensure that your partner has an appropriate level of commitment too, but then don't undervalue the relationship relative to other things in life. Everyone is different, but *to me*, my committed relationship has been far more rewarding than e.g., a more "successful" career would have been. Of course worth noting here that non-men are taught by our society to undervalue their careers & other aspects of their life and sacrifice everything for their partners, which is toxic. I'm not saying "don't value other things" but especially for men, *do* value romantic relationships and be prepared to make decisions that prioritize them over other things, assuming a partner who is comfortable with that commitment and willing to reciprocate.
Okay, this thread is complete for now, until I think of something else that I've missed. I hope this advice is helpful in some way (or at least not harmful). Feel free to chime in if you've got different ideas...
#relationships #love

38% speedup on one of my test setups from a single-character code change.
This bug hiding in libscopehal for ages meant that if you tried to preallocate a buffer to avoid constant resizes, you'd end up reallocating even if the buffer was already the right size. If you're doing this every filter invocation, it adds up.

AcceleratorBuffer: don't reallocate pointlessly if we try to reserve … · ngscopeclient/scopehal@52b550f
…exactly our current size

Repair-R1: Better Test Before Repair
Haichuan Hu, Xiaochen Xie, Quanjun Zhang
https://arxiv.org/abs/2507.22853 https://arxiv.org/pdf/2507.22853

Repair-R1: Better Test Before Repair
APR (Automated Program Repair) aims to automatically locate program defects, generate patches and validate the repairs. Existing techniques for APR are often combined with LLMs (Large Language Models), which leverages the code-related knowledge of LLMs to improve repair effectiveness. Current LLM-based APR methods typically utilize test cases only during the inference stage, adopting an iterative approach that performs repair first and validates it through test execution afterward. This convent…

I was talking to someone yesterday (let's call them A) and they had another "AI" experience, I thought might happen but hadn't heard of before.
They were interacting with an organization and upon asking a specific thing got a very specific answer. Weeks later that organization claimed it had never said what they said and when A showed the email as proof the defense was: Oh yeah, we're an international organization and it's busy right now so the person who sent…

Test-Time Consistency in Vision Language Models
Shih-Han Chou, Shivam Chandhok, James J. Little, Leonid Sigal
https://arxiv.org/abs/2506.22395 https://

Test-Time Consistency in Vision Language Models
Vision-Language Models (VLMs) have achieved impressive performance across a wide range of multimodal tasks, yet they often exhibit inconsistent behavior when faced with semantically equivalent inputs, undermining their reliability and robustness. Recent benchmarks, such as MM-R3, highlight that even state-of-the-art VLMs can produce divergent predictions across semantically equivalent inputs, despite maintaining high average accuracy. Prior work addresses this issue by modifying model architect…