Tootfinder

RADAR: Mechanistic Pathways for Detecting Data Contamination in LLM Evaluation
Ashish Kattamuri, Harshwardhan Fartale, Arpita Vats, Rahul Raja, Ishita Prasad
https://arxiv.org/abs/2510.08931

RADAR: Mechanistic Pathways for Detecting Data Contamination in LLM Evaluation
Data contamination poses a significant challenge to reliable LLM evaluation, where models may achieve high performance by memorizing training data rather than demonstrating genuine reasoning capabilities. We introduce RADAR (Recall vs. Reasoning Detection through Activation Representation), a novel framework that leverages mechanistic interpretability to detect contamination by distinguishing recall-based from reasoning-based model responses. RADAR extracts 37 features spanning surface-level co…

Recall.ai, which offers an API and a desktop SDK for managing meeting recordings and transcripts, raised $38M in a Series B led by Bessemer at a $250M valuation (Paul Gillin/SiliconANGLE)
https://siliconangle.com/2025/09/05/recall-ai-lands-38m-unlock-spoken-…

Recall.ai lands $38M to unlock spoken data at scale - SiliconANGLE
Recall.ai lands $38M to unlock spoken data at scale - SiliconANGLE

Do All Autoregressive Transformers Remember Facts the Same Way? A Cross-Architecture Analysis of Recall Mechanisms
Minyeong Choe, Haehyun Cho, Changho Seo, Hyunil Kim
https://arxiv.org/abs/2509.08778

Do All Autoregressive Transformers Remember Facts the Same Way? A Cross-Architecture Analysis of Recall Mechanisms
Understanding how Transformer-based language models store and retrieve factual associations is critical for improving interpretability and enabling targeted model editing. Prior work, primarily on GPT-style models, has identified MLP modules in early layers as key contributors to factual recall. However, it remains unclear whether these findings generalize across different autoregressive architectures. To address this, we conduct a comprehensive evaluation of factual recall across several model…

Lecture Notes on Verifying Graph Neural Networks
Fran\c{c}ois Schwarzentruber
https://arxiv.org/abs/2510.11617 https://arxiv.org/pdf/2510.11617

Lecture Notes on Verifying Graph Neural Networks
In these lecture notes, we first recall the connection between graph neural networks, Weisfeiler-Lehman tests and logics such as first-order logic and graded modal logic. We then present a modal logic in which counting modalities appear in linear inequalities in order to solve verification tasks on graph neural networks. We describe an algorithm for the satisfiability problem of that logic. It is inspired from the tableau method of vanilla modal logic, extended with reasoning in quantifier-free…

Network Traffic as a Scalable Ethnographic Lens for Understanding University Students' AI Tool Practices
Donghan Hu, Rameen Mahmood, Annabelle David, Danny Yuxing Huang
https://arxiv.org/abs/2510.09763

Network Traffic as a Scalable Ethnographic Lens for Understanding University Students' AI Tool Practices
AI-driven applications have become woven into students' academic and creative workflows, influencing how they learn, write, and produce ideas. Gaining a nuanced understanding of these usage patterns is essential, yet conventional survey and interview methods remain limited by recall bias, self-presentation effects, and the underreporting of habitual behaviors. While ethnographic methods offer richer contextual insights, they often face challenges of scale and reproducibility. To bridge this gap…

Dream to Recall: Imagination-Guided Experience Retrieval for Memory-Persistent Vision-and-Language Navigation
Yunzhe Xu, Yiyuan Pan, Zhe Liu
https://arxiv.org/abs/2510.08553 htt…

Dream to Recall: Imagination-Guided Experience Retrieval for Memory-Persistent Vision-and-Language Navigation
Vision-and-Language Navigation (VLN) requires agents to follow natural language instructions through environments, with memory-persistent variants demanding progressive improvement through accumulated experience. Existing approaches for memory-persistent VLN face critical limitations: they lack effective memory access mechanisms, instead relying on entire memory incorporation or fixed-horizon lookup, and predominantly store only environmental observations while neglecting navigation behavioral …

After typhoon 🌀 Yagi in 2024, the north of #Vietnam faced yet another unprecedented storm. #climatechange
People take hours and hours to get to their families, they swim, the strand.

Associative Memory Model with Neural Networks: Memorizing multiple images with one neuron
Hiroshi Inazawa
https://arxiv.org/abs/2510.06542 https://arxiv.or…

Associative Memory Model with Neural Networks: Memorizing multiple images with one neuron
This paper presents a neural network model (associative memory model) for memory and recall of images. In this model, only a single neuron can memorize multi-images and when that neuron is activated, it is possible to recall all the memorized images at the same time. The system is composed of a single cluster of numerous neurons, referred to as the "Cue Ball," and multiple neural network layers, collectively called the "Recall Net." One of the features of this model is that several different im…

🌟 Astrocytes are superstars in the game of long-term memory
#memory

Astrocytes are superstars in the game of long-term memory
Why are we able to recall only some of our past experiences? A new study led by Jun Nagai at the RIKEN Center for Brain Science in Japan has an answer. Surprisingly, it turns out that the brain cells responsible for stabilizing memories aren't neurons. Rather, they are astrocytes, a type of glial cell that is usually thought of as a role player in the game of learning and memory.

Federated Computation of ROC and PR Curves
Xuefeng Xu, Graham Cormode
https://arxiv.org/abs/2510.04979 https://arxiv.org/pdf/2510.04979

Federated Computation of ROC and PR Curves
Receiver Operating Characteristic (ROC) and Precision-Recall (PR) curves are fundamental tools for evaluating machine learning classifiers, offering detailed insights into the trade-offs between true positive rate vs. false positive rate (ROC) or precision vs. recall (PR). However, in Federated Learning (FL) scenarios, where data is distributed across multiple clients, computing these curves is challenging due to privacy and communication constraints. Specifically, the server cannot access raw …

As failures go, attempting to recall the past is like trying to grasp
the meaning of existence. Both make one feel like a baby clutching at
a basketball: one's palms keep sliding off.
-- Joseph Brodsky

Eagles' A.J. Brown: 'I don't recall' meeting with Jalen Hurts and Saquon Barkley https://www.nytimes.com/athletic/6706567/2025/10/10/a-j-brown-jalen-hurts-saquon-barkley-meeting/

Do you recall back when a three button mouse was a luxury? Like, before scroll wheel, an actual third button? (Doesn't apply to Apple users.)
#PC #nostalgia

Recall Gabor Communication Theory and Joint Time-Frequency Analysis
Xiang-Gen Xia
https://arxiv.org/abs/2509.02724 https://arxiv.org/pdf/2509.02724

Recall Gabor Communication Theory and Joint Time-Frequency Analysis
In this article, we first briefly recall Gabor's communication theory and then Gabor transform and expansion, and also its connection with joint time frequency analysis.

Yikes. I got this notification in the post today. I wonder how complex replacing a starter motor is?
BMW Recalls 196,000 Cars in U.S. Over Fire Risk — Starter Motor Defect
https://www.bmwblog.com/2025/09/26/bmw-recall-starter-motor-fire-risk-2025/

Not to imply any disrespect, just struggling with an old brain vs the collapse of Internet Search, but I can't shake this sense of recalling that, long ago, #JaneGoodall shot someone, a poacher or such, can't recall exactly, maybe it wasnt even Jane, but Search refuses to understand ant permutations of my prompts.

Five principles to protect the human knowledge ecosystem 3/5

Filtered Approximate Nearest Neighbor Search: A Unified Benchmark and Systematic Experimental Study [Experiment, Analysis & Benchmark]
Jiayang Shi, Yuzheng Cai, Weiguo Zheng
https://arxiv.org/abs/2509.07789

Filtered Approximate Nearest Neighbor Search: A Unified Benchmark and Systematic Experimental Study [Experiment, Analysis & Benchmark]
For a given dataset $\mathcal{D}$ and structured label $f$, the goal of Filtered Approximate Nearest Neighbor Search (FANNS) algorithms is to find top-$k$ points closest to a query that satisfy label constraints, while ensuring both recall and QPS (Queries Per Second). In recent years, many FANNS algorithms have been proposed. However, the lack of a systematic investigation makes it difficult to understand their relative strengths and weaknesses. Additionally, we found that: (1) FANNS algorithm…

UniSearch: Rethinking Search System with a Unified Generative Architecture
Jiahui Chen, Xiaoze Jiang, Zhibo Wang, Quanzhi Zhu, Junyao Zhao, Feng Hu, Kang Pan, Ao Xie, Maohua Pei, Zhiheng Qin, Hongjing Zhang, Zhixin Zhai, Xiaobo Guo, Runbin Zhou, Kefeng Wang, Mingyang Geng, Cheng Chen, Jingshan Lv, Yupeng Huang, Xiao Liang, Han Li
https://a…

UniSearch: Rethinking Search System with a Unified Generative Architecture
Modern search systems play a crucial role in facilitating information acquisition. Traditional search engines typically rely on a cascaded architecture, where results are retrieved through recall, pre-ranking, and ranking stages. The complexity of designing and maintaining multiple modules makes it difficult to achieve holistic performance gains. Recent advances in generative recommendation have motivated the exploration of unified generative search as an alternative. However, existing approach…

Estimating Social Network Models with Link Misclassification
Arthur Lewbel, Xi Qu, Xun Tang
https://arxiv.org/abs/2509.07343 https://arxiv.org/pdf/2509.073…

Estimating Social Network Models with Link Misclassification
We propose an adjusted 2SLS estimator for social network models when reported binary network links are misclassified (some zeros reported as ones and vice versa) due, e.g., to survey respondents' recall errors, or lapses in data input. We show misclassification adds new sources of correlation between the regressors and errors, which makes all covariates endogenous and invalidates conventional estimators. We resolve these issues by constructing a novel estimator of misclassification rates and us…

Radioactive Cesium gives the shrimp an extra tangy-ness, and it can give you a warm feeling inside for the rest of your life. 🦐🍤😋🤤🥵🤯💀
But we don’t need no stinkin’ food inspectors, do we doge/maga; thank god for musk/45.
https://time.com/7310679/fda-shrimp-radioactive…

FDA Warns Public Not to Consume Potentially Radioactive Shrimp
The FDA warned about certain frozen shrimp products sold at Walmart due to possible contamination with Cesium-137, a radioactive isotope.

A Recall-First CNN for Sleep Apnea Screening from Snoring Audio
Anushka Mallick, Afiya Noorain, Ashwin Menon, Ashita Solanki, Keertan Balaji
https://arxiv.org/abs/2510.00052 htt…

A Recall-First CNN for Sleep Apnea Screening from Snoring Audio
Sleep apnea is a serious sleep-related breathing disorder that is common and can impact health if left untreated. Currently the traditional method for screening and diagnosis is overnight polysomnography. Polysomnography is expensive and takes a lot of time, and is not practical for screening large groups of people. In this paper, we explored a more accessible option, using respiratory audio recordings to spot signs of apnea.We utilized 18 audio files.The approach involved converting breathing …

Why Can't I See My Clusters? A Precision-Recall Approach to Dimensionality Reduction Validation
Diede P. M. van der Hoorn, Alessio Arleo, Fernando V. Paulovich
https://arxiv.org/abs/2509.04222

Why Can't I See My Clusters? A Precision-Recall Approach to Dimensionality Reduction Validation
Dimensionality Reduction (DR) is widely used for visualizing high-dimensional data, often with the goal of revealing expected cluster structure. However, such a structure may not always appear in the projections. Existing DR quality metrics assess projection reliability (to some extent) or cluster structure quality, but do not explain why expected structures are missing. Visual Analytics solutions can help, but are often time-consuming due to the large hyperparameter space. This paper addresses…

Neural and behavioral reinstatement jointly reflect retrieval of narrative events https://www.nature.com/articles/s41467-025-62375-9 by @…

Neural and behavioral reinstatement jointly reflect retrieval of narrative events - Nature Communications
When people recall a movie, their eye movements and brain activity resemble those observed during the viewing. These behavioral and neural reactivations are linked through a common process, likely reflecting the specific internal experiences that emerge in an instance of recall.

Super homology groups of differential forms and vector fields on Euclidean line
Kentaro Mikami, Tadayoshi Mizutani
https://arxiv.org/abs/2509.04698 https://

Super homology groups of differential forms and vector fields on Euclidean line
The arXiv:2105.09738 claims several stuffs. In particular, we recall the following two. (1) Vector fields and differential forms become a Lie superalgebra structure for each manifold. (2) For an n-dimensional Euclidean space, vector fields and differential forms with polynomial coefficients become a double weighted Lie uperalgebra. By using Euler vector field, the Betti numbers are 0 except the last one if the primary weight and the secondary weight are different. Now, a simple question arises:…

Four years ago, SSR was started to make the great highway permanently car-free. Now, sunset dunes is one of SF's most popular parks. Like the embarcadero and central freeway removals, there may be more fights ahead, but we're not going back.
The recall shows a few things:
-Reactionary politics is inherently unsustainable; the recall-loving, pro-car base that engardio relied on to win in '22 immediately turned on him the minute he did his only visionary act
-Initia…

Conjectural Positivity for Pontryagin Product in Equivariant K-theory of Loop Groups
Shrawan Kumar
https://arxiv.org/abs/2510.07689 https://arxiv.org/pdf/2…

Conjectural Positivity for Pontryagin Product in Equivariant K-theory of Loop Groups
Let $G$ be a connected simply-connected simple algebraic group over $\mathbb{C}$ and let $T$ be a maximal torus, $B\supset T$ a Borel subgroup and $K$ a maximal compact subgroup. Then, the product in the (algebraic) based loop group $Ω(K)$ gives rise to a comultiplication in the topological $T$-equivariant $K$-ring $K_T^{top}(Ω(K))$. Recall that $Ω(K)$ is identified with the affine Grassmannian $\mathcal{X}$ (of $G$) and hence we get a comultiplication in $ K_T^{top}(\mathcal{X})$. Dualizing…

Memory-QA: Answering Recall Questions Based on Multimodal Memories
Hongda Jiang, Xinyuan Zhang, Siddhant Garg, Rishab Arora, Shiun-Zu Kuo, Jiayang Xu, Christopher Brossman, Yue Liu, Aaron Colak, Ahmed Aly, Anuj Kumar, Xin Luna Dong
https://arxiv.org/abs/2509.18436

Memory-QA: Answering Recall Questions Based on Multimodal Memories
We introduce Memory-QA, a novel real-world task that involves answering recall questions about visual content from previously stored multimodal memories. This task poses unique challenges, including the creation of task-oriented memories, the effective utilization of temporal and location information within memories, and the ability to draw upon multiple memories to answer a recall question. To address these challenges, we propose a comprehensive pipeline, Pensieve, integrating memory-specific …

There are a lot of trade-offs with fossil fuel use that we definitely don't discuss enough. #PiperAlpha was the really big one I recall as a kid, but it was just the latest in a whole succession of oil platform accidents that cost a lot of lives and suffering in the North Sea. And then there was #DeepWaterHorizon.
And that's quite apart from the climate costs.
You wonder how many of these happen that we don't hear about too.
This is a good reminder.
https://www.theguardian.com/lifeandstyle/2025/oct/28/the-piper-alpha-oil-rig-exploded-and-collapsed-and-i-made-a-desperate-175ft-jump-into-the-sea

And where are Labour and E&W Greens MPs?
Scottish, Wales and Irish MPs call on Starmer to recall Parliament and ‘impose sanctions’ on Israel | Morning Star
https://morningstaronline.co.uk/article/scottis…

A little Total Recall, back to 1990, before I doze off.. 😂

PSA: there’s a recall on some Anker powerbanks, check if you have one of them
https://www.cpsc.gov/Recalls/2025/Anker-Power-Banks-Recalled-Due-to-Fire-and-Burn-Hazards-Manufactured-by-Anker-Innovations-1

StateX: Enhancing RNN Recall via Post-training State Expansion
Xingyu Shen, Yingfa Chen, Zhen Leng Thai, Xu Han, Zhiyuan Liu, Maosong Sun
https://arxiv.org/abs/2509.22630 https:…

StateX: Enhancing RNN Recall via Post-training State Expansion
While Transformer-based models have demonstrated remarkable language modeling performance, their high complexities result in high costs when processing long contexts. In contrast, recurrent neural networks (RNNs) such as linear attention and state space models have gained popularity due to their constant per-token complexities. However, these recurrent models struggle with tasks that require accurate recall of contextual information from long contexts, because all contextual information is comp…

@… @… a lot of currentColor. Thankfully I don’t recall having to work much with multi-color icons

> Helldivers 2 takes around 150GB to install on PC - three times larger than on console
We did data duplication on optical drives, but we controlled the layout of the whole disc. Is it also so effective on hard drives? I don't recall duplication for our Xbone launch game, but that was the beginning of mandatory installations to hdd on console so maybe I missed this trend.
A multiplayer game demands faster loads, and this keeps all assets for a map contiguous?

Helldivers 2 developer explains why it's so big to install on PC, and why that won't be changing anytime soon
Helldivers 2 takes up around 150GB to install on PC (three times larger than on console), and the game's deputy technical director has now explained why.

Five principles to protect the human knowledge ecosystem. 2/5

Stafford -- along with Calvin Johnson and Cooper Kupp -- recall incredible passes from the playmaker https://www.espn.com/nfl/story/_/id/45965511/rams-matthew-stafford-pass-60000-yards-best-plays

A little byte swap here, a change to CONFIG_LOGO_LINUX_VGA16 there, badabing badaboom - the framebuffer console has the right colors!
Linux assumes framebuffers to be little endian.
In case they are not, as I recall reading, the client is supposed to take care of conversion.
But what if the client is the (Linux internal) console driver?
I couldn't find a config option, so I hacked a byte swap into the color map helper (there are many similarly named functions btw):

i distinctly recall this post that said something along the lines of "this fediverse shit is easy you can just post some shit like 'who up floofing they foxes' and get 7 billion boosts" but i cannot find it for the life of me despite it (iirc) doing pretty decent numbers
i am beginning to think i dreamt it

COBRA: Multimodal Sensing Deep Learning Framework for Remote Chronic Obesity Management via Wrist-Worn Activity Monitoring
Zhengyang Shen (Department of Electrical and Electronic Engineering, Imperial College London, UK), Bo Gao (Department of Electrical and Electronic Engineering, Imperial College London, UK), Mayue Shi (Department of Electrical and Electronic Engineering, Imperial College London, UK)

COBRA: Multimodal Sensing Deep Learning Framework for Remote Chronic Obesity Management via Wrist-Worn Activity Monitoring
Chronic obesity management requires continuous monitoring of energy balance behaviors, yet traditional self-reported methods suffer from significant underreporting and recall bias, and difficulty in integration with modern digital health systems. This study presents COBRA (Chronic Obesity Behavioral Recognition Architecture), a novel deep learning framework for objective behavioral monitoring using wrist-worn multimodal sensors. COBRA integrates a hybrid D-Net architecture combining U-Net spati…

Generation of Grothendieck topologies, provability and operations on subtoposes
Olivia Caramello, Laurent Lafforgue
https://arxiv.org/abs/2508.21134 https://

Generation of Grothendieck topologies, provability and operations on subtoposes
After reviewing the multiple roles of toposes - as generalized topological spaces, as universal invariants, as categorical analogues of the set-theoretic universe, and as semantic environments for first-order theories - we recall the notion of subtopos and its dual expression: both in terms of Grothendieck topologies and in terms of first-order logic. We emphasize the significance of this duality, which enables the translation of provability problems in first-order logic into problems concernin…

Crosslisted article(s) found for cs.IT. https://arxiv.org/list/cs.IT/new
[1/1]:
- Recall Gabor Communication Theory and Joint Time-Frequency Analysis
Xiang-Gen Xia

Comparison of Speech Tasks in Human Expert and Machine Detection of Parkinson's Disease
Peter Plantinga, Roozbeh Sattari, Karine Marcotte, Carla Di Gironimo, Madeleine Sharp, Liziane Bouvier, Maiya Geddes, Ingrid Verduyckt, \'Etienne de Villers-Sidani, Mirco Ravanelli, Denise Klein
https://arxiv.org/abs/2510.07299

Comparison of Speech Tasks in Human Expert and Machine Detection of Parkinson's Disease
The speech of people with Parkinson's Disease (PD) has been shown to hold important clues about the presence and progression of the disease. We investigate the factors based on which humans experts make judgments of the presence of disease in speech samples over five different speech tasks: phonations, sentence repetition, reading, recall, and picture description. We make comparisons by conducting listening tests to determine clinicians accuracy at recognizing signs of PD from audio alone, and …

RF-DETR for Robust Mitotic Figure Detection: A MIDOG 2025 Track 1 Approach
Piotr Giedziun, Jan So{\l}tysik, Mateusz G\'orczany, Norbert Ropiak, Marcin Przymus, Piotr Krajewski, Jaros{\l}aw Kwiecie\'n, Artur Bartczak, Izabela Wasiak, Mateusz Maniewski
https://arxiv.org/abs/2509.02599

RF-DETR for Robust Mitotic Figure Detection: A MIDOG 2025 Track 1 Approach
Mitotic figure detection in histopathology images remains challenging due to significant domain shifts across different scanners, staining protocols, and tissue types. This paper presents our approach for the MIDOG 2025 challenge Track 1, focusing on robust mitotic figure detection across diverse histological contexts. While we initially planned a two-stage approach combining high-recall detection with subsequent classification refinement, time constraints led us to focus on optimizing a single…

Novelist Ruth Ozeki is so delightful.
I am currently reading her "The Book of Form and Emptiness" and after an eventful first chapter, one of the characters suddenly breaks the third wall and addresses "the book", complaining it hadn't started at the beginning, as it should have. 😂
I can't recall encountering this before. Is there some other book where this happens?

Yet another exponential Hopfield model
Linda Albanese, Andrea Alessandrelli, Adriano Barra, Peter Sollich
https://arxiv.org/abs/2509.06905 https://arxiv.or…

Yet another exponential Hopfield model
We propose and analyze a new variation of the so-called {\em exponential Hopfield model}, a recently introduced family of associative neural networks with unprecedented storage capacity. Our construction is based on a cost function defined through exponentials of standard quadratic loss functions, which naturally favors configurations corresponding to perfect recall. Despite not being a mean-field system, the model admits a tractable mathematical analysis of its dynamics and retrieval propertie…

Native Hybrid Attention for Efficient Sequence Modeling
Jusen Du, Jiaxi Hu, Tao Zhang, Weigao Sun, Yu Cheng
https://arxiv.org/abs/2510.07019 https://arxiv.…

Native Hybrid Attention for Efficient Sequence Modeling
Transformers excel at sequence modeling but face quadratic complexity, while linear attention offers improved efficiency but often compromises recall accuracy over long contexts. In this work, we introduce Native Hybrid Attention (NHA), a novel hybrid architecture of linear and full attention that integrates both intra \& inter-layer hybridization into a unified layer design. NHA maintains long-term context in key-value slots updated by a linear RNN, and augments them with short-term tokens fro…

Temporal Drift in Privacy Recall: Users Misremember From Verbatim Loss to Gist-Based Overexposure Over Time
Haoze Guo, Ziqi Wei
https://arxiv.org/abs/2509.16962 https://

Temporal Drift in Privacy Recall: Users Misremember From Verbatim Loss to Gist-Based Overexposure Over Time
With social media content traversing the different platforms, occasionally resurfacing after periods of time, users are increasingly prone to unintended disclosure resulting from a misremembered acceptance of privacy. Context collapse and interface cues are two factors considered by prior researchers, yet we know less about how time-lapse basically alters recall of past audiences destined for exposure. Likewise, the design space for mitigating this temporal exposure risk remains underexplored. …

I had my annual #Medicare wellness appt this morning. Those of you who partake of this ritual will recall the bit where you’re given a sheet of white paper and told to draw a clock face. I was told to make it look like 5:25. So I drew something like the image here. The young man who gave me this task looked at it and suggested I try again. I said it’s right. He said it’s backwards. I showed him th…

Dense Subgraph Clustering and a New Cluster Ensemble Method
The-Anh Vu-Le, Jo\~ao Alfredo Cardoso Lamy, Tom\'as Alessi, Ian Chen, Minhyuk Park, Elfarouk Harb, George Chacko, Tandy Warnow
https://arxiv.org/abs/2508.17013

Dense Subgraph Clustering and a New Cluster Ensemble Method
We propose DSC-Flow-Iter, a new community detection algorithm that is based on iterative extraction of dense subgraphs. Although DSC-Flow-Iter leaves many nodes unclustered, it is competitive with leading methods and has high-precision and low-recall, making it complementary to modularity-based methods that typically have high recall but lower precision. Based on this observation, we introduce a novel cluster ensemble technique that combines DSC-Flow-Iter with modularity-based clustering, to pr…

The Onion *really* needs to find a new gig.

@… @… legacy freebsd-update does respect $EDITOR.
I can't recall how it behaves when $EDITOR is not set, but I almost certainly did experiment at least once in the past.
It took around seven years for me to figure out the five-ste…

Medicare agency recalling thousands of furloughed staffers for open enrollment https://ground.news/article/cms-recalls-nearly-3-000-employees-to-manage-open-enrollment-amid-shutdown

As a legal journalist covering the Justice Department,
I had never encountered anything quite like my exchange with Lindsey Halligan. Neither had my editor.
Over the last several days, he and I spoke with multiple former federal officials and journalists who cover the justice system.
None could recall a similar instance in which a sitting U.S. attorney reached out to chastise a reporter about matters concerning grand jury testimony in an active case.
After I reach…

High-capacity associative memory in a quantum-optical spin glass
Brendan P. Marsh, David Atri Schuller, Yunpeng Ji, Henry S. Hunt, Surya Ganguli, Sarang Gopalakrishnan, Jonathan Keeling, Benjamin L. Lev
https://arxiv.org/abs/2509.12202

High-capacity associative memory in a quantum-optical spin glass
The Hopfield model describes a neural network that stores memories using all-to-all-coupled spins. Memory patterns are recalled under equilibrium dynamics. Storing too many patterns breaks the associative recall process because frustration causes an exponential number of spurious patterns to arise as the network becomes a spin glass. Despite this, memory recall in a spin glass can be restored, and even enhanced, under quantum-optical nonequilibrium dynamics because spurious patterns can now ser…

Every once in a while, I bike through Rosedale or Forest Hill and recall that wealth stratification in #Toronto is even greater than it seems most elsewhere. If the perpetually broke city raised property taxes by only a percent or two on only these and a few other postal codes, it'd have more money than it'd know what to do with and the residents of those areas wouldn't even notice.

Improved Vehicle Maneuver Prediction using Game Theoretic Priors
Nishant Doshi
https://arxiv.org/abs/2509.21873 https://arxiv.org/pdf/2509.21873

Improved Vehicle Maneuver Prediction using Game Theoretic Priors
Conventional maneuver prediction methods use some sort of classification model on temporal trajectory data to predict behavior of agents over a set time horizon. Despite of having the best precision and recall, these models cannot predict a lane change accurately unless they incorporate information about the entire scene. Level-k game theory can leverage the human-like hierarchical reasoning to come up with the most rational decisions each agent can make in a group. This can be leveraged to mod…

Five principles to protect the human knowledge ecosystem. 2/5

Renormalization of Chern-Simons Wilson Loops via Flux Quantization in Cohomotopy
Hisham Sati, Urs Schreiber
https://arxiv.org/abs/2509.25336 https://arxiv.…

Renormalization of Chern-Simons Wilson Loops via Flux Quantization in Cohomotopy
In the practice of physics model building, the process of renormalization, resummation, and anomaly cancellation is to incrementally repair initially ill-defined Lagrangian quantum field theories. Impressive as this is, one would rather have concisely defined complete theories to begin with, and understand these choices as emergent from fundamental principles. As an instructive example, we recall renormalization choices for Wilson loop observables in abelian Chern-Simons theory. Then we show th…

Does the Skeleton-Recall Loss Really Work?
Devansh Arora, Nitin Kumar, Sukrit Gupta
https://arxiv.org/abs/2508.11374 https://arxiv.org/pdf/2508.11374

Does the Skeleton-Recall Loss Really Work?
Image segmentation is an important and widely performed task in computer vision. Accomplishing effective image segmentation in diverse settings often requires custom model architectures and loss functions. A set of models that specialize in segmenting thin tubular structures are topology preservation-based loss functions. These models often utilize a pixel skeletonization process claimed to generate more precise segmentation masks of thin tubes and better capture the structures that other model…

1. Order two bags of cat litter.
2. The seller doesn't send it on time. No matter, you're not in a hurry.
3. Recall that they still didn't send it. Decide to resign.
4. Turns out they send it just before you resign.
5. The parcels come. Open the box.
6. Discover that the bag is damaged. Take a photo just in case.
7. Carefully take it out and check how damaged it is. Well, it's only ripped at the top, no point in returning it over that.
8. Notice they've sent the wrong litter (you've ordered non-scented) 🤦.
9. Call them asking whether there's a point in opening the other box. They say it's going to be the wrong litter too, just make a return request and attach photos.
10. You do that, requesting money return. They request you to make damage protocol from delivery company.
11. You do that. Maybe you'd agree to a 20 PLN discount for next shopping instead of returning two bags of the wrong cat litter?
12. They finally accept the return. And they prepare the return… for one bag 🤦.
So I was supposed to work less today. Instead, I've wasted a lot of time on dealing with damaged parcel and the vendor. And a lot of tape, so it wouldn't fall apart on the return journey.

Sources: Foxconn recalled ~300 Chinese engineers from its Indian plant making iPhone parts, replacing them with Taiwanese staff, its second such move in months (Bloomberg)
https://www.bloomberg.com/news/articles/2025-08-23/fo…

Safeguarding Patient Trust in the Age of AI: Tackling Health Misinformation with Explainable AI
Sueun Hong, Shuojie Fu, Ovidiu Serban, Brianna Bao, James Kinross, Francesa Toni, Guy Martin, Uddhav Vaghela
https://arxiv.org/abs/2509.04052

Safeguarding Patient Trust in the Age of AI: Tackling Health Misinformation with Explainable AI
AI-generated health misinformation poses unprecedented threats to patient safety and healthcare system trust globally. This white paper presents an explainable AI framework developed through the EPSRC INDICATE project to combat medical misinformation while enhancing evidence-based healthcare delivery. Our systematic review of 17 studies reveals the urgent need for transparent AI systems in healthcare. The proposed solution demonstrates 95% recall in clinical evidence retrieval and integrates no…

'Two Hall of Fame quarterbacks just battling it out': A mini-oral history of 2015 Superdome Duel https://www.espn.com/nfl/story/_/id/46479338/mini-oral-history-2015-superdome-duel-new-orleans-saint…

Joint Channel and Semantic-aware Grouping for Effective Collaborative Edge Inference
Mateus P. Mota, Mattia Merluzzi, Emilio Calvanese Strinati
https://arxiv.org/abs/2510.02191 …

Joint Channel and Semantic-aware Grouping for Effective Collaborative Edge Inference
We focus on collaborative edge inference over wireless, which enables multiple devices to cooperate to improve inference performance in the presence of corrupted data. Exploiting a key-query mechanism for selective information exchange (or, group formation for collaboration), we recall the effect of wireless channel impairments in feature communication. We argue and show that a disjoint approach, which only considers either the semantic relevance or channel state between devices, performs poorl…

An introduction to conifold transitions
Tristan C. Collins
https://arxiv.org/abs/2509.01002 https://arxiv.org/pdf/2509.01002…

An introduction to conifold transitions
These lecture notes introduce conifold transitions between complex threefolds with trivial canonical bundle from the differential geometric point of view, and with a particular view towards aspects of mathematical physics and string theory. The lecture notes are aimed at beginning graduate students and non-experts, emphasizing explicit calculations and examples. After a brief introduction in Section 1, we recall some basic facts about Calabi-Yau manifolds in Section 2. Section 3 studies the con…

China says Xiaomi will issue a software update for 115K SU7 EVs to fix assisted driving safety issues, its second recall-related fix since the 2024 launch (Reuters)
https://www.reuters.com/technology/chinas-xiaomi-f…

Totally unsurprising in that industry. Also do recall Sam Altman recently proudly proclaiming OpenAI as a ready and waiting police informer, and Google deciding its primary role is national security.
MIT now reports it need not be you confessing your crimes against the regime to AI, it could be your therapist.
https://www.tumblr.com/teledyn/794863366355206144/totally-unsurprising-in-that-industry-also-do?source=share

MMSearch-Plus: A Simple Yet Challenging Benchmark for Multimodal Browsing Agents
Xijia Tao, Yihua Teng, Xinxing Su, Xinyu Fu, Jihao Wu, Chaofan Tao, Ziru Liu, Haoli Bai, Rui Liu, Lingpeng Kong
https://arxiv.org/abs/2508.21475

MMSearch-Plus: A Simple Yet Challenging Benchmark for Multimodal Browsing Agents
Large multimodal language models (MLLMs) are increasingly deployed as web agents, yet many multimodal browsing benchmarks can be solved by shallow, fixed workflows that lean on high-recall image search and nearby text-masking the genuinely multimodal challenges of fine-grained visual reasoning, provenance verification, and long-horizon tool use. We introduce MMSearch-Plus, a benchmark of 311 tasks that highly demand multimodal understanding while preserving the difficulty profile of strong text…

Do you recall how kids play "shop", trading imaginary wares for imaginary money?
Capitalism is just like that, except it's adults doing that, with other adults watching them and trading even more imaginary money to have a share in their imaginary profits.
#AntiCapitalism

Improving Narrative Classification and Explanation via Fine Tuned Language Models
Rishit Tyagi, Rahul Bouri, Mohit Gupta
https://arxiv.org/abs/2509.04077 https://

Improving Narrative Classification and Explanation via Fine Tuned Language Models
Understanding covert narratives and implicit messaging is essential for analyzing bias and sentiment. Traditional NLP methods struggle with detecting subtle phrasing and hidden agendas. This study tackles two key challenges: (1) multi-label classification of narratives and sub-narratives in news articles, and (2) generating concise, evidence-based explanations for dominant narratives. We fine-tune a BERT model with a recall-oriented approach for comprehensive narrative detection, refining predi…

OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search
Ben Chen, Xian Guo, Siyuan Wang, Zihan Liang, Yue Lv, Yufei Ma, Xinlong Xiao, Bowen Xue, Xuxin Zhang, Ying Yang, Huangyu Dai, Xing Xu, Tong Zhao, Mingcan Peng, XiaoYang Zheng, Cong Zhang, Qihang Zhao, Yuqing Ding, Chenyi Lei, Wenwu Ou, Han Li
https://

OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search
Traditional e-commerce search systems employ multi-stage cascading architectures (MCA) that progressively filter items through recall, pre-ranking, and ranking stages. While effective at balancing computational efficiency with business conversion, these systems suffer from fragmented computation and optimization objective collisions across stages, which ultimately limit their performance ceiling. To address these, we propose \textbf{OneSearch}, the first industrial-deployed end-to-end generativ…

The Command Line GUIde: Graphical Interfaces from Man Pages via AI
Saketh Ram Kasibatla, Kiran Medleri Hiremath, Raven Rothkopf, Sorin Lerner, Haijun Xia, Brian Hempel
https://arxiv.org/abs/2510.01453 …

The Command Line GUIde: Graphical Interfaces from Man Pages via AI
Although birthed in the era of teletypes, the command line shell survived the graphical interface revolution of the 1980's and lives on in modern desktop operating systems. The command line provides access to powerful functionality not otherwise exposed on the computer, but requires users to recall textual syntax and carefully scour documentation. In contrast, graphical interfaces let users organically discover and invoke possible actions through widgets and menus. To better expose the power of…

Spatio-Temporal Directed Graph Learning for Account Takeover Fraud Detection
Mohsen Nayebi Kerdabadi, William Andrew Byron, Xin Sun, Amirfarrokh Iranitalab
https://arxiv.org/abs/2509.20339

Spatio-Temporal Directed Graph Learning for Account Takeover Fraud Detection
Account Takeover (ATO) fraud poses a significant challenge in consumer banking, requiring high recall under strict latency while minimizing friction for legitimate users. Production systems typically rely on tabular gradient-boosted decision trees (e.g., XGBoost) that score sessions independently, overlooking the relational and temporal structure of online activity that characterizes coordinated attacks and "fraud rings." We introduce ATLAS (Account Takeover Learning Across Spatio-Temporal Dire…

A morning thought: claiming that someone is "bothered" by the intensive smell of perfume is an unfair simplification.
It's not that I "don't like" intensive smells. It's an actual physiological reaction. I recall that a long time ago, I accidentally sat behind a classmate who used a lot of perfume. Took me a while to realize why I was getting dizzy wherever I leaned down to note something (at first I thought that leaning was the problem, not entering the cloud of perfume). Using perfumes is a choice, not getting dizzy over smells.
Same goes for noise. It's not that I "don't like" noise. It's that noise kills my nervous system. It's not that I have a choice.
#ActuallyAutistic

How Do Language Models Compose Functions?
Apoorv Khandelwal, Ellie Pavlick
https://arxiv.org/abs/2510.01685 https://arxiv.org/pdf/2510.01685

How Do Language Models Compose Functions?
While large language models (LLMs) appear to be increasingly capable of solving compositional tasks, it is an open question whether they do so using compositional mechanisms. In this work, we investigate how feedforward LLMs solve two-hop factual recall tasks, which can be expressed compositionally as $g(f(x))$. We first confirm that modern LLMs continue to suffer from the "compositionality gap": i.e. their ability to compute both $z = f(x)$ and $y = g(z)$ does not entail their ability to compu…

When recalling in-context, Transformers are not SSMs
Destiny Okpekpe, Antonio Orvieto
https://arxiv.org/abs/2508.19029 https://arxiv.org/pdf/2508.19029

When recalling in-context, Transformers are not SSMs
Despite the advantageous subquadratic complexity of modern recurrent deep learning models -- such as state-space models (SSMs) -- recent studies have highlighted their potential shortcomings compared to transformers on reasoning and memorization tasks. In this paper, we dive deeper into one of such benchmarks: associative recall (AR), which has been shown to correlate well with language modeling performance, and inspect in detail the effects of scaling and optimization issues in recently propos…

Informed Dataset Selection
Abdullah Abbas, Michael Heep, Theodor Sperle
https://arxiv.org/abs/2509.26448 https://arxiv.org/pdf/2509.26448

Informed Dataset Selection
The selection of datasets in recommender systems research lacks a systematic methodology. Researchers often select datasets based on popularity rather than empirical suitability. We developed the APS Explorer, a web application that im- plements the Algorithm Performance Space (APS) framework for informed dataset selection. The system analyzes 96 datasets using 28 algorithms across three metrics (nDCG, Hit Ratio, Recall) at five K-values. We extend the APS framework with a statistical based cla…

Towards a Holistic and Automated Evaluation Framework for Multi-Level Comprehension of LLMs in Book-Length Contexts
Jiaqi Deng, Yuho Lee, Nicole Hee-Yeon Kim, Hyangsuk Min, Taewon Yun, Minjeong Ban, Kim Yul, Hwanjun Song
https://arxiv.org/abs/2508.19578

Towards a Holistic and Automated Evaluation Framework for Multi-Level Comprehension of LLMs in Book-Length Contexts
We introduce HAMLET, a holistic and automated framework for evaluating the long-context comprehension of large language models (LLMs). HAMLET structures source texts into a three-level key-fact hierarchy at root-, branch-, and leaf-levels, and employs query-focused summarization to evaluate how well models recall and faithfully represent information at each level. To validate the reliability of our fully automated pipeline, we conduct a systematic human study, showing that our automatic evaluat…