Tootfinder

Can LLMs Forecast Internet Traffic from Social Media?
Jonatan Langlet, Mariano Scazzariello, Flavio Luciani, Marta Burocchi, Dejan Kosti\'c, Marco Chiesa
https://arxiv.org/abs/2509.20123

Can LLMs Forecast Internet Traffic from Social Media?
Societal events shape the Internet's behavior. The death of a prominent public figure, a software launch, or a major sports match can trigger sudden demand surges that overwhelm peering points and content delivery networks. Although these events fall outside regular traffic patterns, forecasting systems still rely solely on those patterns and therefore miss these critical anomalies. Thus, we argue for socio-technical systems that supplement technical measurements with an active understanding …

When Instructions Multiply: Measuring and Estimating LLM Capabilities of Multiple Instructions Following
Keno Harada, Yudai Yamazaki, Masachika Taniguchi, Edison Marrese-Taylor, Takeshi Kojima, Yusuke Iwasawa, Yutaka Matsuo
https://arxiv.org/abs/2509.21051

When Instructions Multiply: Measuring and Estimating LLM Capabilities of Multiple Instructions Following
As large language models (LLMs) are increasingly applied to real-world scenarios, it becomes crucial to understand their ability to follow multiple instructions simultaneously. To systematically evaluate these capabilities, we introduce two specialized benchmarks for fundamental domains where multiple instructions following is important: Many Instruction-Following Eval (ManyIFEval) for text generation with up to ten instructions, and Style-aware Mostly Basic Programming Problems (StyleMBPP) for…

Matz steals code so we steal code
(these repositories were stolen from their maintainers by Ruby Central)
https://www.ruby-lang.org/en/news/2025/10/17/rubygems-repository-transition/

"""
Customarily, the honour of having liberated hysteria from the ancient myths about a displacement of the uterus goes to Le Pois and Willis. Jean Liebaud, translating or rather adapting Marinello’s work for the seventeenth century, still accepted (with a small number of caveats) the idea of a spontaneous movement of the womb. If it moved, it was “to be more at ease; not that this came about through prudence, nor was it a conscious decision or an animal stimulus, but by a natural instinct, to safeguard health and to have the pleasure of something delectable.” The idea that it could change its place and move around the body, bringing convulsions and spasms everywhere it travelled, had been abandoned, for it was now taken to be ‘tightly held in place’ by the cervix, ligaments, vessels and the sheath of the peritoneum; yet in some senses it could change its location. “The womb therefore, even though it is tightly fixed to the parts that we have described and cannot easily change its place, still manages to roam, making strange, petulant movements around the woman’s body. These diverse movements include ascensions and descents, convulsions, wanderings and prolapses. It can wander up to the liver, spleen, diaphragm, stomach, chest, heart, lung, throat and head.” Physicians of the classical age are more or less unanimous in refusing this explanation.
[…] Yet these analyses were not sufficient to break the theme of an essential link between hysteria and the womb. But the link is now conceived in different terms. It is no longer considered to be the trajectory of a real displacement through the body, but rather a sort of mute propagation through the paths of the organism and its functional proximities. It cannot be said that the seat of the malady has become the brain, nor that thanks to Willis a psychological explanation of hysteria was now possible. But the brain does take on the role of a relay that distributes a malady whose origins are visceral, and the womb brings it on just as the other viscera do. Up until the end of the eighteenth century, and Pinel, the uterus and the womb are still present in the pathology of hysteria, but thanks to a privileged diffusion by the humours and nerves, not because of any particular prestige of their nature.
"""
(Michel Foucault, History of Madness)

Things that were great this weekend: playing copious amounts of FF6. Put in about 24 hours into the game and I'm still not halfway in. Just an awesome game.
Things that weren't a great idea this weekend: Putting in 24 hours to the game. I swear that starting work this morning I could literally see CRT scanlines on my monitors while writing Terraform.
#retrogaming

Day 10: Stacey Mason
Another academic, but this time one of my compatriots; we overlapped at UC Santa Cruz as advisees of Michael Mates, and even collaborated on a Twitch stream called ScholarsPlay for a bit, although we never coauthored any papers. We did chat about our research, and I had many good discussions with her about agency in interactive fiction, a topic we both published on. Her paper "On Games and Links: Extending the Vocabulary of Agency and Immersion in Interactive Narratives" (#20AuthorsNoMen

Topic-Guided Reinforcement Learning with LLMs for Enhancing Multi-Document Summarization
Chuyuan Li, Austin Xu, Shafiq Joty, Giuseppe Carenini
https://arxiv.org/abs/2509.09852 h…

Topic-Guided Reinforcement Learning with LLMs for Enhancing Multi-Document Summarization
A key challenge in Multi-Document Summarization (MDS) is effectively integrating information from multiple sources while maintaining coherence and topical relevance. While Large Language Models have shown impressive results in single-document summarization, their performance on MDS still leaves room for improvement. In this paper, we propose a topic-guided reinforcement learning approach to improve content selection in MDS. We first show that explicitly prompting models with topic labels enhanc…

BioBlue: Notable runaway-optimiser-like LLM failure modes on biologically and economically aligned AI safety benchmarks for LLMs with simplified observation format
Roland Pihlakas, Sruthi Kuriakose
https://arxiv.org/abs/2509.02655

BioBlue: Notable runaway-optimiser-like LLM failure modes on biologically and economically aligned AI safety benchmarks for LLMs with simplified observation format
Relatively many past AI safety discussions have centered around the dangers of unbounded utility maximisation by RL agents, illustrated by scenarios like the "paperclip maximiser" or by specification gaming in general. Unbounded maximisation is problematic for many reasons. We wanted to verify whether these RL runaway optimisation problems are still relevant with LLMs as well. Turns out, strangely, this is indeed clearly the case. The problem is not that the LLMs just lose context or become inc…

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).

LLMs and their Limited Theory of Mind: Evaluating Mental State Annotations in Situated Dialogue
Katharine Kowalyshyn, Matthias Scheutz
https://arxiv.org/abs/2509.02292 https://

LLMs and their Limited Theory of Mind: Evaluating Mental State Annotations in Situated Dialogue
What if large language models could not only infer human mindsets but also expose every blind spot in team dialogue such as discrepancies in the team members' joint understanding? We present a novel, two-step framework that leverages large language models (LLMs) both as human-style annotators of team dialogues to track the team's shared mental models (SMMs) and as automated discrepancy detectors among individuals' mental states. In the first step, an LLM generates annotations by identifying SMM…