Tootfinder

from my link log —
Alan Kay did not invent object-oriented programming.
https://www.hillelwayne.com/post/alan-kay/
saved 2025-05-11 https://

This https://arxiv.org/abs/1809.01751 has been replaced.
link: https://scholar.google.com/scholar?q=a

Misreading EPR: Variations on an Incorrect Theme
Notwithstanding its great influence in modern physics, the EPR thought-experiment has been explained incorrectly a surprising number of times.

On Trustworthy Rule-Based Models and Explanations
Mohamed Siala, Jordi Planes, Joao Marques-Silva
https://arxiv.org/abs/2507.07576 https://

On Trustworthy Rule-Based Models and Explanations
A task of interest in machine learning (ML) is that of ascribing explanations to the predictions made by ML models. Furthermore, in domains deemed high risk, the rigor of explanations is paramount. Indeed, incorrect explanations can and will mislead human decision makers. As a result, and even if interpretability is acknowledged as an elusive concept, so-called interpretable models are employed ubiquitously in high-risk uses of ML and data mining (DM). This is the case for rule-based ML models,…

This https://arxiv.org/abs/2412.06430 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSE_…

Subgraph-Oriented Testing for Deep Learning Libraries
Deep Learning (DL) libraries, such as PyTorch, are widely used for building and deploying DL models on various hardware platforms. Meanwhile, they are found to contain bugs that lead to incorrect calculation results and cause issues like non-convergence training and inaccurate prediction of DL models. Thus, many efforts have been made to test DL libraries and reveal bugs. However, existing DL library testing methods manifest limitations: model-level testing methods cause complexity in fault loc…

Disa: Accurate Learning-based Static Disassembly with Attentions
Peicheng Wang, Monika Santra, Mingyu Liu, Cong Sun, Dongrui Zeng, Gang Tan
https://arxiv.org/abs/2507.07246

Disa: Accurate Learning-based Static Disassembly with Attentions
For reverse engineering related security domains, such as vulnerability detection, malware analysis, and binary hardening, disassembly is crucial yet challenging. The fundamental challenge of disassembly is to identify instruction and function boundaries. Classic approaches rely on file-format assumptions and architecture-specific heuristics to guess the boundaries, resulting in incomplete and incorrect disassembly, especially when the binary is obfuscated. Recent advancements of disassembly ha…

KP-PINNs: Kernel Packet Accelerated Physics Informed Neural Networks
Siyuan Yang, Cheng Song, Zhilu Lai, Wenjia Wang
https://arxiv.org/abs/2506.08563 https…

KP-PINNs: Kernel Packet Accelerated Physics Informed Neural Networks
Differential equations are involved in modeling many engineering problems. Many efforts have been devoted to solving differential equations. Due to the flexibility of neural networks, Physics Informed Neural Networks (PINNs) have recently been proposed to solve complex differential equations and have demonstrated superior performance in many applications. While the L2 loss function is usually a default choice in PINNs, it has been shown that the corresponding numerical solution is incorrect and…

Around 20 years ago, I made my first and only change to something on Wikipedia. It was for an underground rap artist- I just made a correction or two to something that was obviously incorrect, and my information was correct, thinking nothing of it. I looked at it the next day, and a Wikipedia editor or mod or whatever changed it back. That was the last time I ever edited anything there. I still look up things on Wikipedia (with a grain of salt) but that experience really bugged me. The Wikip…

Skip a Layer or Loop it? Test-Time Depth Adaptation of Pretrained LLMs
Ziyue Li, Yang Li, Tianyi Zhou
https://arxiv.org/abs/2507.07996 https://arxiv.org/pdf/2507.07996 https://arxiv.org/html/2507.07996
arXiv:2507.07996v1 Announce Type: new
Abstract: Can a pretrained neural network adapt its architecture to different inputs without any finetuning? Do we need all layers for simple tasks, and are they adequate for challenging tasks? We found that the layers of a pretrained large language model (LLM) can be manipulated as separate modules to build a better and even shallower model customized for each test sample. In particular, each layer from the pretrained model can be skipped/pruned or repeated multiple times as recurrent neural networks (RNN), and stacked with others in arbitrary orders, yielding a chain-of-layers (CoLa) per sample. This compositional space greatly expands the scope of existing works on looped/recurrent pretrained modules, layer pruning, or early-exit networks. We develop a Monte Carlo Tree Search (MCTS) protocol to explore and identify the optimal CoLa for each sample from math and commonsense reasoning benchmarks. Compared to a static model of a fixed depth, CoLa allows shortcut paths (fast thinking), recurrence of the same layer(s) (slow thinking), and combining both, offering more flexible, dynamic architectures for different inputs. We conduct an extensive analysis of the MCTS-optimized CoLa, which leads to two key findings: (1) For >75% of samples with correct predictions by the original LLM, we can find shorter CoLa, suggesting a large space for improving inference efficiency; (2) For >60% of samples with originally incorrect predictions, we can identify CoLa achieving correct predictions, suggesting a large space of performance enhancement. Our results highlight the shortcomings of using a fixed architecture of pre-trained LLMs for inference on different samples and pave the way to unlock the generalization power of test-time depth adaptation.
toXiv_bot_toot

This https://arxiv.org/abs/2411.04215 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_qu…

Sound certification of memory-bounded quantum computers
The rapid advancement of quantum hardware calls for the development of reliable methods to certify its correct functioning. However, existing certification tests often fall short: they either rely on flawless state preparation and measurement or lack soundness guarantees, meaning that they do not rule out incorrect implementations of the target operations by a quantum device. We introduce an approach, which we call quantum system quizzing, for the certification of quantum gates in a practical s…

Adversarial Attack Classification and Robustness Testing for Large Language Models for Code
Yang Liu, Armstrong Foundjem, Foutse Khomh, Heng Li
https://arxiv.org/abs/2506.07942

Adversarial Attack Classification and Robustness Testing for Large Language Models for Code
Large Language Models (LLMs) have become vital tools in software development tasks such as code generation, completion, and analysis. As their integration into workflows deepens, ensuring robustness against vulnerabilities especially those triggered by diverse or adversarial inputs becomes increasingly important. Such vulnerabilities may lead to incorrect or insecure code generation when models encounter perturbed task descriptions, code, or comments. Prior research often overlooks the role of …

Wikipedia editors adopt a policy giving admins the authority to quickly delete AI-generated articles that meet certain criteria, like incorrect citations (Emanuel Maiberg/404 Media)
https://www.404media.co/wikipedia-editors-adopt-speedy-deletion-p…

Wikipedia Editors Adopt ‘Speedy Deletion’ Policy for AI Slop Articles
“The ability to quickly generate a lot of bogus content is problematic if we don't have a way to delete it just as quickly.”

Wikipedia editors adopt a policy giving admins the authority to quickly delete AI-generated articles that meet certain criteria, like incorrect citations (Emanuel Maiberg/404 Media)
https://www.404media.co/wikipedia-editors-adopt-speedy-deletion-p…

Wikipedia Editors Adopt ‘Speedy Deletion’ Policy for AI Slop Articles
“The ability to quickly generate a lot of bogus content is problematic if we don't have a way to delete it just as quickly.”

Where Photons Have Been: Nowhere Without All Components of Their Wavefunctions
R. E. Kastner
https://arxiv.org/abs/2507.06362 https://

Where Photons Have Been: Nowhere Without All Components of Their Wavefunctions
A nested interferometer experiment by Danan et al (2013) is discussed and some ontological implications explored, primarily in the context of time-symmetric interpretations of quantum theory. It is pointed out that photons are supported by all components of their wavefunctions, not selectively truncated "first order" portions of them, and that figures representing both a gap in the photon's path and signals from the cut-off path are incorrect. It is also noted that the Transactional Formulation…

A Refutation of Elmasry's $\tilde{O}(m \sqrt{n})$-Time Algorithm for Single-Source Shortest Paths
Sunny Atalig, Marek Chrobak
https://arxiv.org/abs/2508.04872 https://

A Refutation of Elmasry's $\tilde{O}(m \sqrt{n})$-Time Algorithm for Single-Source Shortest Paths
In this note we examine the recent paper "Breaking the Bellman-Ford Shortest-Path Bound" by Amr Elmasry, where he presents an algorithm for the single-source shortest path problem and claims that its running time complexity is $\tilde{O}(m\sqrt{n})$, where $n$ is the number of vertices and $m$ is the number of edges. We show that his analysis is incorrect, by providing an example of a weighted graph on which the running time of his algorithm is $Ω(mn)$.

As recently as last year I was recommending to my #Math students that if they were having trouble with an elementary algebra expression, they could try using the Google search bar to solve or simplify something. Maybe I should show them this instead.🤦

A tightness criterion for fragmentations
Gabriel Berzunza Ojeda, Cecilia Holmgren, Svante Janson
https://arxiv.org/abs/2507.05102 https://

A tightness criterion for fragmentations
This note presents a simple criterion for the tightness of stochastic fragmentation processes. Our work is motivated by an application to a fragmentation process derived from deleting edges in a conditioned Galton-Watson tree studied by Berzunza-Ojeda and Holmgren (2023). In that paper, while finite-dimensional convergence was established, the claimed functional convergence relied on Lemma 22 of Broutin and Marckert (2016), which is unfortunately incorrect. We show how our results can correct t…

How to use the internet badly:
- I did a search on opening hours for the Scotrail ticket office in my town, got 6 different answers, 2 of them from Scotrail sources
- I tweeted to their twitter acct, telling them abt the variety of answers online, and could they please tell me the correct opening hours
- They tweeted back with what turned out to be incorrect information
- I thanked them and said they should contact the other travel sites and update their info /1

Should we teach vibe coding? Here's why not.
Should AI coding be taught in undergrad CS education?
1/2
I teach undergraduate computer science labs, including for intro and more-advanced core courses. I don't publish (non-negligible) scholarly work in the area, but I've got years of craft expertise in course design, and I do follow the academic literature to some degree. In other words, In not the world's leading expert, but I have spent a lot of time thinking about course design, and consider myself competent at it, with plenty of direct experience in what knowledge & skills I can expect from students as they move through the curriculum.
I'm also strongly against most uses of what's called "AI" these days (specifically, generative deep neutral networks as supplied by our current cadre of techbro). There are a surprising number of completely orthogonal reasons to oppose the use of these systems, and a very limited number of reasonable exceptions (overcoming accessibility barriers is an example). On the grounds of environmental and digital-commons-pollution costs alone, using specifically the largest/newest models is unethical in most cases.
But as any good teacher should, I constantly question these evaluations, because I worry about the impact on my students should I eschew teaching relevant tech for bad reasons (and even for his reasons). I also want to make my reasoning clear to students, who should absolutely question me on this. That inspired me to ask a simple question: ignoring for one moment the ethical objections (which we shouldn't, of course; they're very stark), at what level in the CS major could I expect to teach a course about programming with AI assistance, and expect students to succeed at a more technically demanding final project than a course at the same level where students were banned from using AI? In other words, at what level would I expect students to actually benefit from AI coding "assistance?"
To be clear, I'm assuming that students aren't using AI in other aspects of coursework: the topic of using AI to "help you study" is a separate one (TL;DR it's gross value is not negative, but it's mostly not worth the harm to your metacognitive abilities, which AI-induced changes to the digital commons are making more important than ever).
So what's my answer to this question?
If I'm being incredibly optimistic, senior year. Slightly less optimistic, second year of a masters program. Realistic? Maybe never.
The interesting bit for you-the-reader is: why is this my answer? (Especially given that students would probably self-report significant gains at lower levels.) To start with, [this paper where experienced developers thought that AI assistance sped up their work on real tasks when in fact it slowed it down] (https://arxiv.org/abs/2507.09089) is informative. There are a lot of differences in task between experienced devs solving real bugs and students working on a class project, but it's important to understand that we shouldn't have a baseline expectation that AI coding "assistants" will speed things up in the best of circumstances, and we shouldn't trust self-reports of productivity (or the AI hype machine in general).
Now we might imagine that coding assistants will be better at helping with a student project than at helping with fixing bugs in open-source software, since it's a much easier task. For many programming assignments that have a fixed answer, we know that many AI assistants can just spit out a solution based on prompting them with the problem description (there's another elephant in the room here to do with learning outcomes regardless of project success, but we'll ignore this over too, my focus here is on project complexity reach, not learning outcomes). My question is about more open-ended projects, not assignments with an expected answer. Here's a second study (by one of my colleagues) about novices using AI assistance for programming tasks. It showcases how difficult it is to use AI tools well, and some of these stumbling blocks that novices in particular face.
But what about intermediate students? Might there be some level where the AI is helpful because the task is still relatively simple and the students are good enough to handle it? The problem with this is that as task complexity increases, so does the likelihood of the AI generating (or copying) code that uses more complex constructs which a student doesn't understand. Let's say I have second year students writing interactive websites with JavaScript. Without a lot of care that those students don't know how to deploy, the AI is likely to suggest code that depends on several different frameworks, from React to JQuery, without actually setting up or including those frameworks, and of course three students would be way out of their depth trying to do that. This is a general problem: each programming class carefully limits the specific code frameworks and constructs it expects students to know based on the material it covers. There is no feasible way to limit an AI assistant to a fixed set of constructs or frameworks, using current designs. There are alternate designs where this would be possible (like AI search through adaptation from a controlled library of snippets) but those would be entirely different tools.
So what happens on a sizeable class project where the AI has dropped in buggy code, especially if it uses code constructs the students don't understand? Best case, they understand that they don't understand and re-prompt, or ask for help from an instructor or TA quickly who helps them get rid of the stuff they don't understand and re-prompt or manually add stuff they do. Average case: they waste several hours and/or sweep the bugs partly under the rug, resulting in a project with significant defects. Students in their second and even third years of a CS major still have a lot to learn about debugging, and usually have significant gaps in their knowledge of even their most comfortable programming language. I do think regardless of AI we as teachers need to get better at teaching debugging skills, but the knowledge gaps are inevitable because there's just too much to know. In Python, for example, the LLM is going to spit out yields, async functions, try/finally, maybe even something like a while/else, or with recent training data, the walrus operator. I can't expect even a fraction of 3rd year students who have worked with Python since their first year to know about all these things, and based on how students approach projects where they have studied all the relevant constructs but have forgotten some, I'm not optimistic seeing these things will magically become learning opportunities. Student projects are better off working with a limited subset of full programming languages that the students have actually learned, and using AI coding assistants as currently designed makes this impossible. Beyond that, even when the "assistant" just introduces bugs using syntax the students understand, even through their 4th year many students struggle to understand the operation of moderately complex code they've written themselves, let alone written by someone else. Having access to an AI that will confidently offer incorrect explanations for bugs will make this worse.
To be sure a small minority of students will be able to overcome these problems, but that minority is the group that has a good grasp of the fundamentals and has broadened their knowledge through self-study, which earlier AI-reliant classes would make less likely to happen. In any case, I care about the average student, since we already have plenty of stuff about our institutions that makes life easier for a favored few while being worse for the average student (note that our construction of that favored few as the "good" students is a large part of this problem).
To summarize: because AI assistants introduce excess code complexity and difficult-to-debug bugs, they'll slow down rather than speed up project progress for the average student on moderately complex projects. On a fixed deadline, they'll result in worse projects, or necessitate less ambitious project scoping to ensure adequate completion, and I expect this remains broadly true through 4-6 years of study in most programs (don't take this as an endorsement of AI "assistants" for masters students; we've ignored a lot of other problems along the way).
There's a related problem: solving open-ended project assignments well ultimately depends on deeply understanding the problem, and AI "assistants" allow students to put a lot of code in their file without spending much time thinking about the problem or building an understanding of it. This is awful for learning outcomes, but also bad for project success. Getting students to see the value of thinking deeply about a problem is a thorny pedagogical puzzle at the best of times, and allowing the use of AI "assistants" makes the problem much much worse. This is another area I hope to see (or even drive) pedagogical improvement in, for what it's worth.
1/2

The BBC rejects a White House claim that it took down a Gaza story, after Press Secretary Karoline Leavitt accused the BBC of taking "the word of Hamas" (Barbara Tasch/BBC)
https://www.bbc.com/news/articles/ce814ez7030o

BBC rejects incorrect White House claims on Gaza coverage
The corporation says claims that a story was removed from its website are "completely wrong".

Benford's Curse: Tracing Digit Bias to Numerical Hallucination in LLMs
Jiandong Shao, Yao Lu, Jianfei Yang
https://arxiv.org/abs/2506.01734 https://

Benford's Curse: Tracing Digit Bias to Numerical Hallucination in LLMs
Large Language Models (LLMs) exhibit impressive performance on complex reasoning tasks, yet they frequently fail on basic numerical problems, producing incorrect outputs. Inspired by Benford's Law -- a statistical pattern where lower digits occur more frequently as leading digits -- we hypothesize that the long-tailed digit distributions in web-collected corpora may be learned by LLMs during pretraining, leading to biased numerical generation. To investigate the hypothesis, we first examine whe…

This https://arxiv.org/abs/2412.09840 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDC_…

LAVA: Lifetime-Aware VM Allocation with Learned Distributions and Adaptation to Mispredictions
Scheduling virtual machines (VMs) on hosts in cloud data centers dictates efficiency and is an NP-hard problem with incomplete information. Prior work improved VM scheduling with predicted VM lifetimes. Our work further improves lifetime-aware scheduling using repredictions with lifetime distributions versus one-shot prediction. Our approach repredicts and adjusts VM and host lifetimes when incorrect predictions emerge. We also present novel approaches for defragmentation and regular system mai…

Dear Apple SwiftUI friends, this bug is killing the vibe of our users: FB16257334
Repro for those in the community:
https://github.com/feedback-assistant/reports/issues/594

FB16257334: Inspector view bug · Issue #594 · feedback-assistant/reports
Date: 2024-01-07 Resolution: Open Area: SwiftUI OS: iPadOS 18.1 Type: Incorrect/Unexpected Behavior Keywords: SwiftUI, split view, iPadOS, NavigationSplitView Description Simple NavigationSplitView...

Remember when Jimmy "The Greek" Snyder made those arrogantly racist & fundamentally incorrect comments in the 80s?
He got FIRED from CBS for that.
Your move, #NewYorkTimes.
☑️ The New York Times Just Published Some Bizarre Race Science About Asian Women

The New York Times Just Published Some Bizarre Race Science About Asian Women
The New York Times published an article with a strange and bigoted claim about Asian women and manufacturing at Apple.

Advanced Purrsistent Threat (APT)

Diffusion Models for Increasing Accuracy in Olfaction Sensors and Datasets
Kordel K. France, Ovidiu Daescu
https://arxiv.org/abs/2506.00455 https://…

Diffusion Models for Increasing Accuracy in Olfaction Sensors and Datasets
Robotic odour source localization (OSL) is a critical capability for autonomous systems operating in complex environments. However, current OSL methods often suffer from ambiguities, particularly when robots misattribute odours to incorrect objects due to limitations in olfactory datasets and sensor resolutions. To address this challenge, we introduce a novel machine learning method using diffusion-based molecular generation to enhance odour localization accuracy that can be used by itself or w…

Silent Data Corruption by 10x Test Escapes Threatens Reliable Computing
Subhasish Mitra, Subho Banerjee, Martin Dixon, Rama Govindaraju, Peter Hochschild, Eric X. Liu, Bharath Parthasarathy, Parthasarathy Ranganathan
https://arxiv.org/abs/2508.01786

Silent Data Corruption by 10x Test Escapes Threatens Reliable Computing
Too many defective compute chips are escaping existing manufacturing tests -- at least an order of magnitude more than industrial targets across all compute chip types in data centers. Silent data corruptions (SDCs) caused by test escapes, when left unaddressed, pose a major threat to reliable computing. We present a three-pronged approach to future directions in overcoming test escapes: (a) Quick diagnosis of defective chips directly from system-level incorrect behaviors. Such diagnosis is cri…

Estimating Correctness Without Oracles in LLM-Based Code Generation
Thomas Valentin, Ardi Madadi, Gaetano Sapia, Marcel B\"ohme
https://arxiv.org/abs/2507.00057

Estimating Correctness Without Oracles in LLM-Based Code Generation
Generating code from natural language specifications is one of the most successful applications of Large Language Models (LLMs). Yet, they hallucinate: LLMs produce outputs that may be grammatically correct but are factually incorrect. Without an existing, correct implementation (i.e., an oracle), can we quantify how likely the generated program is correct? In this paper, we propose a measure of incorrectness, called incoherence, that can be estimated efficiently in the absence of an oracle a…

hassediagrams:an R package that generates the Hasse diagram of the layout structure and the restricted layout structure
Damianos Michaelides, Simon T. Bate, Marion J. Chatfield
https://arxiv.org/abs/2507.05949

hassediagrams:an R package that generates the Hasse diagram of the layout structure and the restricted layout structure
With the advent of modern statistical software, complex experimental designs are now routinely employed in many areas of research. Failing to correctly identify the structure of the experimental design can lead to incorrect model selection and misleading inferences. This paper describes the hassediagrams package in R that determines the structure of the design, summarised by the layout structure, and generates a Hasse diagram of the layout structure. By considering the randomisation performed, …

This https://arxiv.org/abs/2505.20982 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_hepp…

On chirality and chiral neutrino oscillations
It has been claimed in a number of publications that neutrinos can exhibit chirality oscillations. In this note we discuss the notion of chirality and show that chiral neutrino oscillations in vacuum do not occur. We argue that the incorrect claims to the contrary resulted from a failure to clearly discriminate between quantum fields, states and wave functions. We also emphasize the role played in the erroneous claims on the possibility of chirality oscillations by the widely spread misconcepti…

Distinguishing Fact from Fiction: Student Traits, Attitudes, and AI Hallucination Detection in Business School Assessment
Canh Thien Dang, An Nguyen
https://arxiv.org/abs/2506.00050

Distinguishing Fact from Fiction: Student Traits, Attitudes, and AI Hallucination Detection in Business School Assessment
As artificial intelligence (AI) becomes integral to the society, the ability to critically evaluate AI-generated content is increasingly vital. On the context of management education, we examine how academic skills, cognitive traits, and AI scepticism influence students' ability to detect factually incorrect AI-generated responses (hallucinations) in a high-stakes assessment at a UK business school (n=211, Year 2 economics and management students). We find that only 20% successfully identified …

This https://arxiv.org/abs/2503.00303 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csHC_…

Leveraging Complementary AI Explanations to Mitigate Misunderstanding in XAI
Artificial intelligence explanations can make complex predictive models more comprehensible. To be effective, however, they should anticipate and mitigate possible misinterpretations, e.g., arising when users infer incorrect information that is not explicitly conveyed. To this end, we propose complementary explanations -- a novel method that pairs explanations to compensate for their respective limitations. A complementary explanation adds insights that clarify potential misconceptions stemming…

Spherically Symmetric Accretion with Self-Gravity: Analytical Formulae and Numerical Validation
Cheng-Liang Jiao, Er-gang Zhao, Liying Zhu, Xiang-dong Shi
https://arxiv.org/abs/2507.02621

Spherically Symmetric Accretion with Self-Gravity: Analytical Formulae and Numerical Validation
Spherically symmetric accretion incorporating self-gravity constitutes a three-point boundary value problem (TPBVP) governed by constraints at the outer boundary, sonic point, and accretor surface. Previous studies have two limitations: either employing an incorrect formula for self-gravity potential in analytical treatments, or introducing additional input parameters in numerical implementations to circumvent solving the full TPBVP. To address these issues, we present a self-consistent TPBVP f…

Approximation Algorithms for Scheduling Crowdsourcing Tasks in Mobile Social Networks
Chi-Yeh Chen
https://arxiv.org/abs/2508.04159 https://arxiv.org/pdf/2…

Approximation Algorithms for Scheduling Crowdsourcing Tasks in Mobile Social Networks
This paper addresses the scheduling problem in mobile social networks. We begin by proving that the approximation ratio analysis presented in the paper by Zhang \textit{et al.} (IEEE Transactions on Mobile Computing, 2025) is incorrect, and we provide the correct analysis results. Furthermore, when the required service time for a task exceeds the total contact time between the requester and the crowd worker, we demonstrate that the approximation ratio of the Largest-Ratio-First task scheduling …

Bidirectional Age of Incorrect Information: A Performance Metric for Status Updates in Virtual Dynamic Environments
Chiara Schiavo, Manuele Favero, Alessandro Buratto, Leonardo Badia
https://arxiv.org/abs/2507.13312

Bidirectional Age of Incorrect Information: A Performance Metric for Status Updates in Virtual Dynamic Environments
Virtual dynamic environments (VDEs) such as the Metaverse and digital twins (DTs) require proper representation of the interacting entities to map their characteristics within the simulated or augmented space. Keeping these representations accurate and up-to-date is crucial for seamless interaction and system reliability. In this paper, we propose bidirectional age of incorrect information (BAoII) to address this aspect. BAoII quantifies the time-dependent penalty paid by an entity in a VDE due…

Minority Representation in Network Rankings: Methods for Estimation, Testing, and Fairness
Hui Shen, Peter W. MacDonald, Eric D. Kolaczyk
https://arxiv.org/abs/2507.01136

Minority Representation in Network Rankings: Methods for Estimation, Testing, and Fairness
Networks, composed of nodes and their connections, are widely used to model complex relationships across various fields. Centrality metrics often inform decisions such as identifying key nodes or prioritizing resources. However, networks frequently suffer from missing or incorrect edges, which can systematically centrality-based decisions and distort the representation of certain protected groups. To address this issue, we introduce a formal definition of minority representation, measured as th…

I'm seeing some really awful, low effort "may be..." ALT text recently. Clearly generated by an automatic process rather than by a human.
Is bad* alt text worse than no alt text?
*completely incorrect, and misleading
#Accessibility #a11y

Experimenting with Permutation Wordle
Aurora Hiveley
https://arxiv.org/abs/2506.23452 https://arxiv.org/pdf/2506.23452

Experimenting with Permutation Wordle
Consider a game of permutation wordle in which a player attempts to guess a secret permutation of length $n$ in as few guesses as possible. In each round, the guessing player is told which indices of their guessed permutation are correct. How can we optimize the player's strategy? Samuel Kutin and Lawren Smithline (arXiv:2408.00903) propose a strategy called "cyclic shift" in which all incorrect entries are shifted one index to the right in successive guesses, and they conjecture its optimality…

"Deep down in his heart, I believe Trump knows he’s an incompetent a**hole."
Dear Mr King. You are incorrect about the Heart ❤️ part

Classification with Reject Option: Distribution-free Error Guarantees via Conformal Prediction
Johan Hallberg Szabadv\'ary, Tuwe L\"ofstr\"om, Ulf Johansson, Cecilia S\"onstr\"od, Ernst Ahlberg, Lars Carlsson
https://arxiv.org/abs/2506.21802

Classification with Reject Option: Distribution-free Error Guarantees via Conformal Prediction
Machine learning (ML) models always make a prediction, even when they are likely to be wrong. This causes problems in practical applications, as we do not know if we should trust a prediction. ML with reject option addresses this issue by abstaining from making a prediction if it is likely to be incorrect. In this work, we formalise the approach to ML with reject option in binary classification, deriving theoretical guarantees on the resulting error rate. This is achieved through conformal pred…

A strange political story that somehow seems strangely relevant today.
https://apple.news/AZzxjXuHrQNuJ_10aeK3DgA

It is very funny when you get blocked for sharing a "Comparison of Android-based Operating Systems" that I didn't make, and if you think anything is factually incorrect with the comparison chart, you can contribute to it.
https://eylenburg.github.io/android_comparison.htm

#Python world be like:
"Oh, hi, we wrote a new library implementing this spec."
"Hey, it looks like it doesn't conform to the spec, it doesn't pass the examples from it."
"Oh, you're right, we'll fix it ASAP."
…and that was over 3 years ago.
And yet projects keep adding a dependency on this library which has a single "pre-alpha" release 3.5 years ago and whose very first bug report points out it's incorrect.

The EU says it will investigate whether KKR provided incorrect or misleading information in its €22B acquisition of Telecom Italia's fixed-line network (Foo Yun Chee/Reuters)
https://www.reuters.com/legal/litigation/e

Scalability, Availability, Reproducibility and Extensibility in Islamic Database Systems
Umar Siddiqui, Habiba Youssef, Adel Sabour, Mohamed Ali
https://arxiv.org/abs/2507.22384

Scalability, Availability, Reproducibility and Extensibility in Islamic Database Systems
With the widespread of software systems and applications that serve the Islamic knowledge domain, several concerns arise. Authenticity and accuracy of the databases that back up these systems are questionable. With the excitement that some software developers and amateur researchers may have, false statements and incorrect claims may be made around numerical signs or miracles in the Quran. Reproducibility of these claims may not be addressed by the people making such claims. Moreover, with the …

Self-ensemble: Mitigating Confidence Distortion for Large Language Models
Zicheng Xu, Guanchu Wang, Guangyao Zheng, Yu-Neng Chuang, Alexander Szalay, Xia Hu, Vladimir Braverman
https://arxiv.org/abs/2506.01951

Self-ensemble: Mitigating Confidence Distortion for Large Language Models
Although Large Language Models (LLMs) perform well in general fields, they exhibit a confidence distortion problem on multi-choice question-answering (MCQA), particularly as the number of answer choices increases. Specifically, on MCQA with many choices, LLMs suffer from under-confidence in correct predictions and over-confidence in incorrect ones, leading to a substantially degraded performance. To solve this problem, we propose Self-ensemble in this work. Our method splits the choices into se…

"Trump says power plants don't add to air pollution. Climate scientists say it’s 'nonsensical'"
#US #USA #America

Climate scientists hit back on Trump plan to nix air pollution rules
More than a dozen climate scientists said the Trump administration’s assertions were factually incorrect.

SynC: Synthetic Image Caption Dataset Refinement with One-to-many Mapping for Zero-shot Image Captioning
Si-Woo Kim, MinJu Jeon, Ye-Chan Kim, Soeun Lee, Taewhan Kim, Dong-Jin Kim
https://arxiv.org/abs/2507.18616

SynC: Synthetic Image Caption Dataset Refinement with One-to-many Mapping for Zero-shot Image Captioning
Zero-shot Image Captioning (ZIC) increasingly utilizes synthetic datasets generated by text-to-image (T2I) models to mitigate the need for costly manual annotation. However, these T2I models often produce images that exhibit semantic misalignments with their corresponding input captions (e.g., missing objects, incorrect attributes), resulting in noisy synthetic image-caption pairs that can hinder model training. Existing dataset pruning techniques are largely designed for removing noisy text in…

A heads-up for #Interrail in Denmark: The Rail Planner app seems to have some incorrect data at the moment. The ICL train I just took between Kolding and Copenhagen is missing from the app. As always, trust bahn.de or the DB Navigator app more.

This https://arxiv.org/abs/2410.05605 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSE_…

CodeDPO: Aligning Code Models with Self Generated and Verified Source Code
Code generation models have shown significant potential for programming tasks. However, existing training methods like supervised fine-tuning face key limitations: they do not effectively teach models to prioritize correct over incorrect solutions in ambiguous situations, nor do they effectively optimize the runtime efficiency of the generated code. To address these challenges, we propose CodeDPO, a framework that integrates preference learning into code generation to improve two key code prefe…

Beyond Accuracy: How AI Metacognitive Sensitivity improves AI-assisted Decision Making
ZhaoBin Li, Mark Steyvers
https://arxiv.org/abs/2507.22365 https://a…

Beyond Accuracy: How AI Metacognitive Sensitivity improves AI-assisted Decision Making
In settings where human decision-making relies on AI input, both the predictive accuracy of the AI system and the reliability of its confidence estimates influence decision quality. We highlight the role of AI metacognitive sensitivity -- its ability to assign confidence scores that accurately distinguish correct from incorrect predictions -- and introduce a theoretical framework for assessing the joint impact of AI's predictive accuracy and metacognitive sensitivity in hybrid decision-making s…

I noticed that OpenStreetMaps somehow lost Lake Michigan, so I went investigating. Apparently back in April somebody accidentally changed it from a "lake" to a "school" and it's taking months to be fixed across all regions/renderers.
https:/…

The Rayleigh Criterion: Resolution Limits of Astronomical Periodograms
V. Ramirez Delgado, J. S. Caicedo Vivas, S. Dodson-Robinson, C. Haley
https://arxiv.org/abs/2506.20864

The Rayleigh Criterion: Resolution Limits of Astronomical Periodograms
The Rayleigh criterion determines the resolution limit of a periodogram, which is the minimum frequency separation required to barely resolve two sinusoids. Failing to consider the Rayleigh criterion may result in incorrect interpretations of long-period signals or spurious claims that two closely spaced periodogram peaks represent two distinct physical processes. We demonstrate how applying the Rayleigh criterion can help observers avoid false positive detections caused by uneven observing cad…

Impact of Detector Calibration Accuracy on Black Hole Spectroscopy
Mallika R. Sinha, Ling Sun, Sizheng Ma
https://arxiv.org/abs/2506.15979 https://

Impact of Detector Calibration Accuracy on Black Hole Spectroscopy
Systematic errors in detector calibration can bias signal analyses and potentially lead to incorrect interpretations suggesting violations of general relativity. In this study, we investigate how calibration systematics affect black hole (BH) spectroscopy, a technique that uses the quasinormal modes (QNMs) emitted during the ringdown phase of gravitational waves (GWs) to study remnant BHs formed in compact binary coalescences. We simulate a series of physically motivated, tunable calibration er…

R4ec: A Reasoning, Reflection, and Refinement Framework for Recommendation Systems
Hao Gu, Rui Zhong, Yu Xia, Wei Yang, Chi Lu, Peng Jiang, Kun Gai
https://arxiv.org/abs/2507.17249

R4ec: A Reasoning, Reflection, and Refinement Framework for Recommendation Systems
Harnessing Large Language Models (LLMs) for recommendation systems has emerged as a prominent avenue, drawing substantial research interest. However, existing approaches primarily involve basic prompt techniques for knowledge acquisition, which resemble System-1 thinking. This makes these methods highly sensitive to errors in the reasoning path, where even a small mistake can lead to an incorrect inference. To this end, in this paper, we propose $R^{4}$ec, a reasoning, reflection and refinement…

Optimal Parallel Algorithms for Convex Hulls in 2D and 3D under Noisy Primitive Operations
Michael T. Goodrich, Vinesh Sridhar
https://arxiv.org/abs/2506.17507

Optimal Parallel Algorithms for Convex Hulls in 2D and 3D under Noisy Primitive Operations
In the noisy primitives model, each primitive comparison performed by an algorithm, e.g., testing whether one value is greater than another, returns the incorrect answer with random, independent probability p < 1/2 and otherwise returns a correct answer. This model was first applied in the context of sorting and searching, and recent work by Eppstein, Goodrich, and Sridhar extends this model to sequential algorithms involving geometric primitives such as orientation and sidedness tests. However…

You know it's bad when the DeArrow title for a video straight up has "Video has lots of incorrect information" on the end.

A conformal basis for cosmology with energy conservation
J. M. Greben
https://arxiv.org/abs/2508.00948 https://arxiv.org/pdf/2508.00948

A conformal basis for cosmology with energy conservation
In the standard FRW formalism, the scale factor is assumed to describe the expansion of the universe. However, by examining empty space with a positive cosmological constant (i.e., a de Sitter space), we find that this assumption is incorrect. When described in conformal time, the associated conformal metric exhibits a big bang singularity where the effective vacuum energy density diverges, dominating the early universe. The corresponding FRW scale factor decreases after the big bang, so that i…

Classification of Multi-Parametric Body MRI Series Using Deep Learning
Boah Kim, Tejas Sudharshan Mathai, Kimberly Helm, Peter A. Pinto, Ronald M. Summers
https://arxiv.org/abs/2506.15182

Classification of Multi-Parametric Body MRI Series Using Deep Learning
Multi-parametric magnetic resonance imaging (mpMRI) exams have various series types acquired with different imaging protocols. The DICOM headers of these series often have incorrect information due to the sheer diversity of protocols and occasional technologist errors. To address this, we present a deep learning-based classification model to classify 8 different body mpMRI series types so that radiologists read the exams efficiently. Using mpMRI data from various institutions, multiple deep lea…

@… Incorrect, we only think we are because most of us haven’t spent time in Switzerland.

Investigating Hallucination in Conversations for Low Resource Languages
Amit Das, Md. Najib Hasan, Souvika Sarkar, Zheng Zhang, Fatemeh Jamshidi, Tathagata Bhattacharya, Nilanjana Raychawdhury, Dongji Feng, Vinija Jain, Aman Chadha
https://arxiv.org/abs/2507.22720

Investigating Hallucination in Conversations for Low Resource Languages
Large Language Models (LLMs) have demonstrated remarkable proficiency in generating text that closely resemble human writing. However, they often generate factually incorrect statements, a problem typically referred to as 'hallucination'. Addressing hallucination is crucial for enhancing the reliability and effectiveness of LLMs. While much research has focused on hallucinations in English, our study extends this investigation to conversational data in three languages: Hindi, Farsi, and Mandari…

I get the desire for live coding interviews, you can't just take people's word that they know how to do it.
But what's the point in throwing Advent of Code style problems at me and interrupting a naive or incorrect approach before I even start implementing it? Let me write unoptimized code for you and then make it better! I'm not going to write the perfect implementation first try for every problem, but I can show you my process and prove I can write _some_ code.
…

YATE: The Role of Test Repair in LLM-Based Unit Test Generation
Michael Konstantinou, Renzo Degiovanni, Jie M. Zhang, Mark Harman, Mike Papadakis
https://arxiv.org/abs/2507.18316

YATE: The Role of Test Repair in LLM-Based Unit Test Generation
Recent advances in automated test generation utilises language models to produce unit tests. While effective, language models tend to generate many incorrect tests with respect to both syntax and semantics. Although such incorrect tests can be easily detected and discarded, they constitute a "missed opportunity" -- if fixed, they are often valuable as they directly add testing value (they effectively target the underlying program logic to be tested) and indirectly form good seeds for generating…

ASCD: Attention-Steerable Contrastive Decoding for Reducing Hallucination in MLLM
Yujun Wang, Jinhe Bi, Yunpu Ma, Soeren Pirk
https://arxiv.org/abs/2506.14766

ASCD: Attention-Steerable Contrastive Decoding for Reducing Hallucination in MLLM
Multimodal Large Language Model (MLLM) often suffer from hallucinations. They over-rely on partial cues and generate incorrect responses. Recently, methods like Visual Contrastive Decoding (VCD) and Instruction Contrastive Decoding (ICD) have been proposed to mitigate hallucinations by contrasting predictions from perturbed or negatively prefixed inputs against original outputs. In this work, we uncover that methods like VCD and ICD fundamentally influence internal attention dynamics of the mod…

A rediscovery of stiff pentmodes. A comment on "High bulk modulus pentamodes: the three-dimensional metal water"
Graeme W. Milton
https://arxiv.org/abs/2507.15014

A rediscovery of stiff pentmodes. A comment on "High bulk modulus pentamodes: the three-dimensional metal water"
We bring attention to the fact that the claim of Brambilla et.al. [Extreme Mechanics Letters 74 (2025) 102267; arXiv:2406.14502] of discovering a novel design for pentamode materials is incorrect. Back in 2016 Briane, Harutyunyan and myself [Mathematics and Mechanics of Complex Systems 5 (2016) 41--94; arXiv:1606.03305] designed a class of stiff pentamodes, that include the high bulk modulus pentamodes of Brambilla et.al.

Reasoning about Uncertainty: Do Reasoning Models Know When They Don't Know?
Zhiting Mei, Christina Zhang, Tenny Yin, Justin Lidard, Ola Shorinwa, Anirudha Majumdar
https://arxiv.org/abs/2506.18183

Reasoning about Uncertainty: Do Reasoning Models Know When They Don't Know?
Reasoning language models have set state-of-the-art (SOTA) records on many challenging benchmarks, enabled by multi-step reasoning induced using reinforcement learning. However, like previous language models, reasoning models are prone to generating confident, plausible responses that are incorrect (hallucinations). Knowing when and how much to trust these models is critical to the safe deployment of reasoning models in real-world applications. To this end, we explore uncertainty quantification…

FRED: Financial Retrieval-Enhanced Detection and Editing of Hallucinations in Language Models
Likun Tan, Kuan-Wei Huang, Kevin Wu
https://arxiv.org/abs/2507.20930 https://

FRED: Financial Retrieval-Enhanced Detection and Editing of Hallucinations in Language Models
Hallucinations in large language models pose a critical challenge for applications requiring factual reliability, particularly in high-stakes domains such as finance. This work presents an effective approach for detecting and editing factually incorrect content in model-generated responses based on the provided context. Given a user-defined domain-specific error taxonomy, we construct a synthetic dataset by inserting tagged errors into financial question-answering corpora and then fine-tune fou…

Towards Formal Verification of LLM-Generated Code from Natural Language Prompts
Aaron Councilman, David Fu, Aryan Gupta, Chengxiao Wang, David Grove, Yu-Xiong Wang, Vikram Adve
https://arxiv.org/abs/2507.13290

Towards Formal Verification of LLM-Generated Code from Natural Language Prompts
In the past few years LLMs have emerged as a tool that can aid programmers by taking natural language descriptions and generating code based on it. However, LLMs often generate incorrect code that users need to fix and the literature suggests users often struggle to detect these errors. In this work we seek to offer formal guarantees of correctness to LLM generated code; such guarantees could improve the experience of using AI Code Assistants and potentially enable natural language programming …

Is Your Automated Software Engineer Trustworthy?
Noble Saji Mathews, Meiyappan Nagappan
https://arxiv.org/abs/2506.17812 https://arxi…

Is Your Automated Software Engineer Trustworthy?
Large Language Models (LLMs) are being increasingly used in software engineering tasks, with an increased focus on bug report resolution over the past year. However, most proposed systems fail to properly handle uncertain or incorrect inputs and outputs. Existing LLM-based tools and coding agents respond to every issue and generate a patch for every case, even when the input is vague or their own output is incorrect. There are no mechanisms in place to abstain when confidence is low. This leads…

Hard Contacts with Soft Gradients: Refining Differentiable Simulators for Learning and Control
Anselm Paulus, A. Ren\'e Geist, Pierre Schumacher, V\'it Musil, Georg Martius
https://arxiv.org/abs/2506.14186

Hard Contacts with Soft Gradients: Refining Differentiable Simulators for Learning and Control
Contact forces pose a major challenge for gradient-based optimization of robot dynamics as they introduce jumps in the system's velocities. Penalty-based simulators, such as MuJoCo, simplify gradient computation by softening the contact forces. However, realistically simulating hard contacts requires very stiff contact settings, which leads to incorrect gradients when using automatic differentiation. On the other hand, using non-stiff settings strongly increases the sim-to-real gap. We analyze …

On a few pitfalls in KL divergence gradient estimation for RL
Yunhao Tang, R\'emi Munos
https://arxiv.org/abs/2506.09477 https://…

On a few pitfalls in KL divergence gradient estimation for RL
We point out a few pitfalls in implementing gradient estimation for KL divergence in RL training for LLM, as seen in a number of open source projects and papers. The first major pitfall is to differentiate through the KL estimate as loss functions to minimize KL divergence. We show that such implementations are generally incorrect and do not produce the desired KL gradient. Secondly, we show that some implementations do not account for the sequential nature of the estimation problem and produce…

Students' Reliance on AI in Higher Education: Identifying Contributing Factors
Griffin Pitts, Neha Rani, Weedguet Mildort, Eva-Marie Cook
https://arxiv.org/abs/2506.13845

Students' Reliance on AI in Higher Education: Identifying Contributing Factors
The increasing availability and use of artificial intelligence (AI) tools in educational settings has raised concerns about students' overreliance on these technologies. Overreliance occurs when individuals accept incorrect AI-generated recommendations, often without critical evaluation, leading to flawed problem solutions and undermining learning outcomes. This study investigates potential factors contributing to patterns of AI reliance among undergraduate students, examining not only overreli…

LaSM: Layer-wise Scaling Mechanism for Defending Pop-up Attack on GUI Agents
Zihe Yan, Zhuosheng Zhang
https://arxiv.org/abs/2507.10610 https://

LaSM: Layer-wise Scaling Mechanism for Defending Pop-up Attack on GUI Agents
Graphical user interface (GUI) agents built on multimodal large language models (MLLMs) have recently demonstrated strong decision-making abilities in screen-based interaction tasks. However, they remain highly vulnerable to pop-up-based environmental injection attacks, where malicious visual elements divert model attention and lead to unsafe or incorrect actions. Existing defense methods either require costly retraining or perform poorly under inductive interference. In this work, we systemati…

DREAM: On hallucinations in AI-generated content for nuclear medicine imaging
Menghua Xia, Reimund Bayerlein, Yanis Chemli, Xiaofeng Liu, Jinsong Ouyang, Georges El Fakhri, Ramsey D. Badawi, Quanzheng Li, Chi Liu
https://arxiv.org/abs/2506.13995

DREAM: On hallucinations in AI-generated content for nuclear medicine imaging
Artificial intelligence-generated content (AIGC) has shown remarkable performance in nuclear medicine imaging (NMI), offering cost-effective software solutions for tasks such as image enhancement, motion correction, and attenuation correction. However, these advancements come with the risk of hallucinations, generating realistic yet factually incorrect content. Hallucinations can misrepresent anatomical and functional information, compromising diagnostic accuracy and clinical trust. This paper …

Comments on Exploring Quantum Statistics for Dirac and Majorana Neutrinos using Spinor-Helicity technique (arXiv:2507.07180 [hep-ph])
C. S. Kim, M. V. N. Murthy, Dibyakrupa Sahoo
https://arxiv.org/abs/2507.14463

Comments on Exploring Quantum Statistics for Dirac and Majorana Neutrinos using Spinor-Helicity technique (arXiv:2507.07180 [hep-ph])
We give our comments on Ref. [1](arXiv:2507.07180) which critiques our idea of exploring quantum statistics to distinguish between Dirac and Majorana neutrinos proposed in some of our earlier works. We show that the ad-hoc symmetrization of the Dirac case amplitude square advocated in Eqs. (16) and (35) of [1], have no physical basis and it leads to violation of lepton number in the the standard model. Therefore, we consider this symmetrization by hand as incorrect and misleading. Moreover, we …

Mitigating Geospatial Knowledge Hallucination in Large Language Models: Benchmarking and Dynamic Factuality Aligning
Shengyuan Wang, Jie Feng, Tianhui Liu, Dan Pei, Yong Li
https://arxiv.org/abs/2507.19586

Mitigating Geospatial Knowledge Hallucination in Large Language Models: Benchmarking and Dynamic Factuality Aligning
Large language models (LLMs) possess extensive world knowledge, including geospatial knowledge, which has been successfully applied to various geospatial tasks such as mobility prediction and social indicator prediction. However, LLMs often generate inaccurate geospatial knowledge, leading to geospatial hallucinations (incorrect or inconsistent representations of geospatial information) that compromise their reliability. While the phenomenon of general knowledge hallucination in LLMs has been w…

TTrace: Lightweight Error Checking and Diagnosis for Distributed Training
Haitian Jiang, Shaowei Zhu, Zhen Zhang, Zhenyu Song, Xinwei Fu, Zhen Jia, Yida Wang, Jinyang Li
https://arxiv.org/abs/2506.09280

TTrace: Lightweight Error Checking and Diagnosis for Distributed Training
Distributed training is essential for scaling the training of large neural network models, such as large language models (LLMs), across thousands of GPUs. However, the complexity of distributed training programs makes them particularly prone to silent bugs, which do not produce explicit error signal but lead to incorrect training outcome. Effectively detecting and localizing such silent bugs in distributed training is challenging. Common debugging practice using metrics like training loss or gr…

Criteria-Based LLM Relevance Judgments
Naghmeh Farzi, Laura Dietz
https://arxiv.org/abs/2507.09488 https://arxiv.org/pdf/2507.09488…

Criteria-Based LLM Relevance Judgments
Relevance judgments are crucial for evaluating information retrieval systems, but traditional human-annotated labels are time-consuming and expensive. As a result, many researchers turn to automatic alternatives to accelerate method development. Among these, Large Language Models (LLMs) provide a scalable solution by generating relevance labels directly through prompting. However, prompting an LLM for a relevance label without constraints often results in not only incorrect predictions but also…

Text-Aware Image Restoration with Diffusion Models
Jaewon Min, Jin Hyeon Kim, Paul Hyunbin Cho, Jaeeun Lee, Jihye Park, Minkyu Park, Sangpil Kim, Hyunhee Park, Seungryong Kim
https://arxiv.org/abs/2506.09993

Text-Aware Image Restoration with Diffusion Models
Image restoration aims to recover degraded images. However, existing diffusion-based restoration methods, despite great success in natural image restoration, often struggle to faithfully reconstruct textual regions in degraded images. Those methods frequently generate plausible but incorrect text-like patterns, a phenomenon we refer to as text-image hallucination. In this paper, we introduce Text-Aware Image Restoration (TAIR), a novel restoration task that requires the simultaneous recovery of…

The Ethics of Generative AI in Anonymous Spaces: A Case Study of 4chan's /pol/ Board
Parth Gaba, Emiliano De Cristofaro
https://arxiv.org/abs/2506.14191

The Ethics of Generative AI in Anonymous Spaces: A Case Study of 4chan's /pol/ Board
This paper presents a characterization of AI-generated images shared on 4chan, examining how this anonymous online community is (mis-)using generative image technologies. Through a methodical data collection process, we gathered 900 images from 4chan's /pol/ (Politically Incorrect) board, which included the label "/mwg/" (memetic warfare general), between April and July 2024, identifying 66 unique AI-generated images. The analysis reveals concerning patterns in the use of this technology, with …

COIN: Uncertainty-Guarding Selective Question Answering for Foundation Models with Provable Risk Guarantees
Zhiyuan Wang, Jinhao Duan, Qingni Wang, Xiaofeng Zhu, Tianlong Chen, Xiaoshuang Shi, Kaidi Xu
https://arxiv.org/abs/2506.20178

COIN: Uncertainty-Guarding Selective Question Answering for Foundation Models with Provable Risk Guarantees
Uncertainty quantification (UQ) for foundation models is essential to identify and mitigate potential hallucinations in automatically generated text. However, heuristic UQ approaches lack formal guarantees for key metrics such as the false discovery rate (FDR) in selective prediction. Previous work adopts the split conformal prediction (SCP) framework to ensure desired coverage of admissible answers by constructing prediction sets, but these sets often contain incorrect candidates, limiting the…

Is Your Model Risk ALARP? Evaluating Prospective Safety-Critical Applications of Complex Models
Domenic Di Francesco, Alan Forrest, Fiona McGarry, Nicholas Hall, Adam Sobey
https://arxiv.org/abs/2507.10817

Is Your Model Risk ALARP? Evaluating Prospective Safety-Critical Applications of Complex Models
The increasing availability of advanced computational modelling offers new opportunities to improve safety, efficacy, and emissions reductions. Application of complex models to support engineering decisions has been slow in comparison to other sectors, reflecting the higher consequence of unsafe applications. Adopting a complex model introduces a \emph{model risk}, namely the expected consequence of incorrect or otherwise unhelpful outputs. This should be weighed against the prospective benefit…

An Empirical Study of Bugs in Data Visualization Libraries
Weiqi Lu, Yongqiang Tian, Xiaohan Zhong, Haoyang Ma, Zhenyang Xu, Shing-Chi Cheung, Chengnian Sun
https://arxiv.org/abs/2506.15084

An Empirical Study of Bugs in Data Visualization Libraries
Data visualization (DataViz) libraries play a crucial role in presentation, data analysis, and application development, underscoring the importance of their accuracy in transforming data into visual representations. Incorrect visualizations can adversely impact user experience, distort information conveyance, and influence user perception and decision-making processes. Visual bugs in these libraries can be particularly insidious as they may not cause obvious errors like crashes, but instead mis…

The Curious Case of Factuality Finetuning: Models' Internal Beliefs Can Improve Factuality
Benjamin Newman, Abhilasha Ravichander, Jaehun Jung, Rui Xin, Hamish Ivison, Yegor Kuznetsov, Pang Wei Koh, Yejin Choi
https://arxiv.org/abs/2507.08371

The Curious Case of Factuality Finetuning: Models' Internal Beliefs Can Improve Factuality
Language models are prone to hallucination - generating text that is factually incorrect. Finetuning models on high-quality factual information can potentially reduce hallucination, but concerns remain; obtaining factual gold data can be expensive and training on correct but unfamiliar data may potentially lead to even more downstream hallucination. What data should practitioners finetune on to mitigate hallucinations in language models? In this work, we study the relationship between the factu…

Potemkin Understanding in Large Language Models
Marina Mancoridis, Bec Weeks, Keyon Vafa, Sendhil Mullainathan
https://arxiv.org/abs/2506.21521 https://arxiv.org/pdf/2506.21521 https://arxiv.org/html/2506.21521
arXiv:2506.21521v1 Announce Type: new
Abstract: Large language models (LLMs) are regularly evaluated using benchmark datasets. But what justifies making inferences about an LLM's capabilities based on its answers to a curated set of questions? This paper first introduces a formal framework to address this question. The key is to note that the benchmarks used to test LLMs -- such as AP exams -- are also those used to test people. However, this raises an implication: these benchmarks are only valid tests if LLMs misunderstand concepts in ways that mirror human misunderstandings. Otherwise, success on benchmarks only demonstrates potemkin understanding: the illusion of understanding driven by answers irreconcilable with how any human would interpret a concept. We present two procedures for quantifying the existence of potemkins: one using a specially designed benchmark in three domains, the other using a general procedure that provides a lower-bound on their prevalence. We find that potemkins are ubiquitous across models, tasks, and domains. We also find that these failures reflect not just incorrect understanding, but deeper internal incoherence in concept representations.
toXiv_bot_toot