28081999.me

Record 27042026 · captured 2026-08-25

Monday, 27 April 2026

0 tracked subjects, 25 discussions, 25 papers. This record is frozen and will not change.

Complete record JSON

← 2026-04-26 Monday, 27 April 2026 2026-04-28 →

Frontier 25

What the people building things argued about, from Hacker News.

  1. 990points

    Microsoft and OpenAI end their exclusive and revenue-sharing deal

    845 comments

  2. 773points

    Talkie: a 13B vintage language model from 1930

    331 comments

  3. 767points

    GitHub Copilot is moving to usage-based billing

    553 comments

  4. 724points

    Men who stare at walls

    337 comments

  5. 694points

    Is my blue your blue? (2024)

    469 comments

  6. 600points

    4TB of voice samples just stolen from 40k AI contractors at Mercor

    226 comments

  7. 501points

    Noctua releases official 3D CAD models for its cooling fans

    108 comments

  8. 451points

    Pgbackrest is no longer being maintained

    232 comments

  9. 402points

    China blocks Meta's acquisition of AI startup Manus

    338 comments

  10. 393points

    Show HN: OSS Agent I built topped the TerminalBench on Gemini-3-flash-preview

    148 comments

  11. 360points

    You can beat the binary search

    166 comments

  12. 360points

    Dutch central bank ditches AWS and chooses Lidl for European Cloud

    148 comments

  13. 358points

    Quarkdown – Markdown with Superpowers

    143 comments

  14. 339points

    Super ZSNES – GPU Powered SNES Emulator

    116 comments

  15. 338points

    GitHub is having issues now

    123 comments

  16. 334points

    To my students

    199 comments

  17. 304points

    “Why not just use Lean?”

    209 comments

  18. 300points

    TurboQuant: A first-principles walkthrough

    66 comments

  19. 279points

    The Prompt API

    144 comments

  20. 272points

    Networking changes coming in macOS 27

    237 comments

  21. 271points

    The woes of sanitizing SVGs

    109 comments

  22. 257points

    US Supreme Court reviews police use of cell location data

    158 comments

  23. 253points

    FDA approves first gene therapy for treatment of genetic hearing loss

    103 comments

  24. 253points

    Easyduino: Open Source PCB Devboards for KiCad

    44 comments

  25. 234points

    United Wizards of the Coast

    218 comments

Research 25

Papers submitted to arXiv cs.AI that day.

  1. arXiv2112.02604

    PSI: A Benchmark for Human Interpretation and Response in Traffic Interactions

    Accurately modeling pedestrian intention and understanding driver decision-making processes are critical for the development of safe and socially aware autonomous driving systems. We introduce PSI, a benchmark dataset that captures the dynamic evolution of pedestrian crossing intentions from the driver's perspective, enriched with human textual explanations that reflect the reasoning behind intention estimation a

    Jing, Taotao, Chen, Tina, Tian, Renran, Chen, Yaobin et al.

  2. arXiv2209.06865

    Sketch of a novel approach to a neural model

    We present an account of neuroplasticity with respect to cell-internal processing pathways in relation to membrane and synaptic plasticity. We think traditional synapse-centric, weight-based models of memorization are not sufficient or adequate to capture the complexity of neuroplasticity. In these accounts, the model is a network of neurons connected by adaptive transmission links. The adaptation of the transmission

    Scheler, Gabriele

  3. arXiv2210.05513

    ViFiCon: Vision and Wireless Association Via Self-Supervised Contrastive Learning

    We introduce ViFiCon, a self-supervised contrastive scheme which learns a cross-modal association between vision and wireless modalities. Specifically, the system uses pedestrian data collected from RGB-D camera footage and WiFi Fine Time Measurements (FTM) from a user's smartphone device. Depth data from RGB-D (vision domain) is inherently linked with an observable pedestrian, but FTM data (wireless domain) is a

    Meegan, Nicholas, Liu, Hansi, Cao, Bryan Bo, Alali, Abrar et al.

  4. arXiv2211.16327

    On the Power of Foundation Models

    With infinitely many high-quality data points, infinite computational power, an infinitely large foundation model with a perfect training algorithm and guaranteed zero generalization error on the pretext task, can the model be used for everything? This question cannot be answered by the existing theory of representation, optimization or generalization, because the issues they mainly investigate are assumed to be none

    Yuan, Yang

  5. arXiv2305.00931

    Explanation through Reward Model Reconciliation using POMDP Tree Search

    As artificial intelligence (AI) algorithms are increasingly used in mission-critical applications, promoting user-trust of these systems will be essential to their success. Ensuring users understand the models over which algorithms reason promotes user trust. This work seeks to reconcile differences between the reward model that an algorithm uses for online partially observable Markov decision (POMDP) planning and th

    Kraske, Benjamin D., Saksena, Anshu, Buczak, Anna L., Sunberg, Zachary N.

  6. arXiv2405.00577

    Discovering robust biomarkers of psychiatric disorders from resting-state functional MRI via graph neural networks: A systematic review

    Graph neural networks (GNN) have emerged as a popular tool for modelling functional magnetic resonance imaging (fMRI) datasets. Many recent studies have reported significant improvements in disorder classification performance via more sophisticated GNN designs and highlighted salient features that could be potential biomarkers of the disorder. However, existing methods of evaluating their robustness are often limited

    Chan, Yi Hao, Girish, Deepank, Gupta, Sukrit, Xia, Jing et al.

  7. arXiv2405.10138

    PL-MTEB: Polish Massive Text Embedding Benchmark

    In this paper, we introduce the Polish Massive Text Embedding Benchmark (PL-MTEB), a comprehensive benchmark for text embeddings in the Polish language. PL-MTEB comprises 30 diverse NLP tasks across five categories: classification, clustering, pair classification, information retrieval, and semantic text similarity. Within the scope of this work, we added 12 new Polish-language tasks to MTEB based on existing dataset

    Poświata, Rafał, Dadas, Sławomir, Perełkiewicz, Michał

  8. arXiv2406.14111

    Expander Hierarchies for Normalized Cuts on Graphs

    Expander decompositions of graphs have significantly advanced the understanding of many classical graph problems and led to numerous fundamental theoretical results. However, their adoption in practice has been hindered due to their inherent intricacies and large hidden factors in their asymptotic running times. Here, we introduce the first practically efficient algorithm for computing expander decompositions and the

    Hanauer, Kathrin, Henzinger, Monika, Münk, Robin, Räcke, Harald et al.

  9. arXiv2407.08750

    Online Distributional Regression

    Large-scale streaming data are common in modern machine learning applications and have led to the development of online learning algorithms. Many fields, such as supply chain management, weather and meteorology, energy markets, and finance, have pivoted toward probabilistic forecasting. This results in the need not only for accurate learning of the expected value but also for learning the conditional heteroskedastici

    Hirsch, Simon, Berrisch, Jonathan, Ziel, Florian

  10. arXiv2408.16286

    Near-Optimal Policy Identification in Robust Constrained Markov Decision Processes via Epigraph Form

    Designing a safe policy for uncertain environments is crucial in real-world control systems. However, this challenge remains inadequately addressed within the Markov decision process (MDP) framework. This paper presents the first algorithm guaranteed to identify a near-optimal policy in a robust constrained MDP (RCMDP), where an optimal policy minimizes cumulative cost while satisfying constraints in the worst-case s

    Kitamura, Toshinori, Kozuno, Tadashi, Kumagai, Wataru, Hoshino, Kenta et al.

  11. arXiv2409.18169

    Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey

    Recent research demonstrates that the nascent fine-tuning-as-a-service business model exposes serious safety concerns: fine-tuning with a few harmful data uploaded from the users can compromise the safety alignment of the model. The attack, known as harmful fine-tuning attack, has generated broad research interests in both academia and industry. In this paper, we first systematically formulate the threat model and ba

    Huang, Tiansheng, Hu, Sihao, Ilhan, Fatih, Tekin, Selim Furkan et al.

  12. arXiv2410.13713

    CrystalX: High-accuracy Crystal Structure Analysis Using Deep Learning

    Atomic structure analysis of crystalline materials is a paramount endeavor in both chemical and material sciences. This sophisticated technique necessitates not only a solid foundation in crystallography but also a profound comprehension of the intricacies of the accompanying software, posing a significant challenge in meeting the rigorous daily demands. For the first time, we confront this challenge head-on by harne

    Zheng, Kaipeng, Huang, Weiran, Ouyang, Wanli, Zhong, Han-Sen et al.

  13. arXiv2410.21548

    MultiTok: Variable-Length Tokenization for Efficient LLMs Adapted from LZW Compression

    Large language models have drastically changed the prospects of AI by introducing technologies for more complex natural language processing. However, current methodologies to train such LLMs require extensive resources including but not limited to large amounts of data, expensive machinery, and lengthy training. To solve this problem, this paper proposes a new tokenization method inspired by universal Lempel-Ziv-Welc

    Elias, Noel, Esfahanizadeh, Homa, Kale, Kaan, Vishwanath, Sriram et al.

  14. arXiv2411.07378

    Data-Driven Analysis of AI in Medical Device Software in China: Trends of Deep Learning and Traditional AI Based on Regulatory Data

    Artificial intelligence (AI) in medical device software (MDSW) represents a transformative clinical technology, attracting increasing attention within both the medical community and the regulators. In this study, we leverage a data-driven approach to automatically extract and analyze AI-enabled medical devices (AIMD) from the National Medical Products Administration (NMPA) regulatory database. The continued increase

    Han, Yu, Ceross, Aaron, Ather, Sarim, Bergmann, Jeroen H. M.

  15. arXiv2411.08027

    LLMPhy: Parameter-Identifiable Physical Reasoning Combining Large Language Models and Physics Engines

    Most learning-based approaches to complex physical reasoning sidestep the crucial problem of parameter identification (e.g., mass, friction) that governs scene dynamics, despite its importance in real-world applications such as collision avoidance and robotic manipulation. In this paper, we present LLMPhy, a black-box optimization framework that integrates large language models (LLMs) with physics simulators for phys

    Cherian, Anoop, Corcodel, Radu, Jain, Siddarth, Romeres, Diego

  16. arXiv2412.19780

    Tensor Network Estimation of Distribution Algorithms

    Tensor networks are a tool first employed in the context of many-body quantum physics that now have a wide range of uses across the computational sciences, from numerical methods to machine learning. Methods integrating tensor networks into evolutionary optimization algorithms have appeared in the recent literature. In essence, these methods can be understood as replacing the traditional crossover operation of a gene

    Gardiner, John, Lopez-Piqueres, Javier

  17. arXiv2501.16839

    Flow Matching: Markov Kernels, Stochastic Processes and Transport Plans

    Among generative neural models, flow matching techniques stand out for their simple applicability and good scaling properties. Here, velocity fields of curves connecting a simple latent and a target distribution are learned. Then the corresponding ordinary differential equation can be used to sample from a target distribution, starting in samples from the latent one. This paper reviews from a mathematical point of vi

    Wald, Christian, Steidl, Gabriele

  18. arXiv2501.19277

    On Pareto Optimality for Parametric Choice Bandits

    We study online assortment optimization under stochastic choice when a decision maker simultaneously values cumulative revenue performance and the quality of post-hoc inference on revenue contrasts. We analyze a forced-exploration optimism-in-the-face-of-uncertainty (OFU) scheme that combines two regularized maximum-likelihood estimators: one based on all observations for sequential decision making, and one based onl

    Zuo, Jierui, Qin, Hanzhang

  19. arXiv2502.00955

    Efficient Multi-Agent System Training with Data Influence-Oriented Tree Search

    Monte Carlo Tree Search (MCTS) based methods provide promising approaches for generating synthetic data to enhance the self-training of Large Language Model (LLM) based multi-agent systems (MAS). These methods leverage Q-values to estimate individual agent contributions. However, relying solely on Q-values to identify informative data may misalign with the data synthesis objective, as the focus should be on selecting

    Shi, Wentao, Yu, Zichun, Feng, Fuli, He, Xiangnan et al.

  20. arXiv2502.03698

    How Vulnerable Is My Learned Policy? Universal Adversarial Perturbation Attacks On Modern Behavior Cloning Policies

    Learning from demonstrations is a popular approach to train AI models; however, their vulnerability to adversarial attacks remains underexplored. We present the first systematic study of adversarial attacks, across a range of both classic and recently proposed imitation learning algorithms, including Vanilla Behavior Cloning (Vanilla BC), LSTM-GMM, Implicit Behavior Cloning (IBC), Diffusion Policy (DP), and Vector-Qu

    Kalra, Akansha, Patil, Basavasagar, Tao, Guanhong, Brown, Daniel S.

  21. arXiv2502.16994

    FADE: Why Bad Descriptions Happen to Good Features

    Recent advances in mechanistic interpretability have highlighted the potential of automating interpretability pipelines in analyzing the latent representations within LLMs. While this may enhance our understanding of internal mechanisms, the field lacks standardized evaluation methods for assessing the validity of discovered features. We attempt to bridge this gap by introducing FADE: Feature Alignment to Description

    Puri, Bruno, Jain, Aakriti, Golimblevskaia, Elena, Kahardipraja, Patrick et al.

  22. arXiv2502.17011

    Predicting Liquidity-Aware Bond Yields using Causal GANs and Deep Reinforcement Learning with LLM Evaluation

    Financial bond yield forecasting is challenging due to data scarcity, nonlinear macroeconomic dependencies, and evolving market conditions. In this paper, we propose a novel framework that leverages Causal Generative Adversarial Networks (CausalGANs) and Soft Actor-Critic (SAC) reinforcement learning (RL) to generate high-fidelity synthetic bond yield data for four major bond categories (AAA, BAA, US10Y, Junk). By in

    Walia, Jaskaran Singh, Sinha, Aarush, Saraswat, Naman, Srinivasan, Srinitish et al.

  23. arXiv2503.05231

    Kaiwu: A Multimodal Manipulation Dataset and Framework for Robot Learning and Human-Robot Interaction

    Cutting-edge robot learning techniques including foundation models and imitation learning from humans all pose huge demands on large-scale and high-quality datasets which constitute one of the bottleneck in the general intelligent robot fields. This paper presents the Kaiwu multimodal dataset to address the missing real-world synchronized multimodal data problems in the sophisticated assembling scenario,especially wi

    Jiang, Shuo, Li, Haonan, Ren, Ruochen, Zhou, Yanmin et al.

  24. arXiv2503.12507

    Segment Any-Quality Images with Generative Latent Space Enhancement

    Despite their success, Segment Anything Models (SAMs) experience significant performance drops on severely degraded, low-quality images, limiting their effectiveness in real-world scenarios. To address this, we propose GleSAM, which utilizes Generative Latent space Enhancement to boost robustness on low-quality images, thus enabling generalization across various image qualities. Specifically, we adapt the concept of

    Guo, Guangqian, Guo, Yong, Yu, Xuehui, Li, Wenbo et al.

  25. arXiv2503.21435

    Graph-to-Vision: Multi-graph Understanding and Reasoning using Vision-Language Models

    Recent advances in Vision-Language Models (VLMs) have shown promising capabilities in interpreting visualized graph data, offering a new perspective for graph-structured reasoning beyond traditional Graph Neural Networks (GNNs). However, existing studies focus primarily on single-graph reasoning, leaving the critical challenge of multi-graph joint reasoning underexplored. In this work, we introduce the first comprehe

    Ai, Qihang, Li, Ruizhou, Wang, Menghui, Jiang, Haiyun

On this calendar day 18

Notable events recorded on this day and month across all years.

  1. 2012year

    Unknown perpetrators carried out a series of four bombings in Dnipropetrovsk, Ukraine.

  2. 2011year

    During the most active day of the 2011 Super Outbreak 218 tornadoes killed 317 people across 16 U.S. states.

  3. 2005year

    The Airbus A380, the world's largest passenger airliner, made its maiden flight from Toulouse, France.

  4. 1997year

    American spree killer Andrew Cunanan murdered Jeffrey Trail, a former friend and a Gulf War veteran, who became the first victim of Cunanan's killing spree.

  5. 1993year

    Thirty people died, including players and staff of the Zambia football team and the crew, in a plane crash en route to play a World Cup qualifying match against Senegal.

  6. 1985year

    The black-ball final, one of the most famous snooker matches in history, began between Steve Davis and Dennis Taylor.

  7. 1967year

    The Expo 67 world's fair opened in Montreal, with 62 nations participating and more than 50 million visitors ultimately attending.

  8. 1965year

    Indonesia–Malaysia confrontation: British forces repelled a surprise Indonesian attack on a base at Plaman Mapu in Sarawak.

  9. 1961year

    Prime Minister Milton Margai led the Sierra Leone Colony and Protectorate to independence from the United Kingdom.

  10. 1949year

    In response to the treatment of Lorenzo Gamboa under the White Australia policy, the Philippine House of Representatives passed a bill banning Australians from the country.

  11. 1945year

    World War II: The photograph Raising the Flag on the Three-Country Cairn (pictured) was taken after German troops withdrew to Norway at the end of the Lapland War.

  12. 1936year

    The trilogy The Adventure Girls (second book pictured), by the children's author Clair Blank, was copyrighted by A. L. Burt.

  13. 1911year

    The Second Canton Uprising took place but was suppressed in Guangzhou, China.

  14. 1904year

    Chris Watson became the first prime minister of Australia from the Australian Labor Party.

  15. 1848year

    The French National Assembly under Victor Schœlcher proclaims the abolition of slavery within France and it's colonies following the February Revolution.

  16. 1810year

    Ludwig van Beethoven composed his Bagatelle No. 25 in A minor, better known as "Für Elise" , one of his most popular compositions.

  17. 1805year

    First Barbary War: U.S. Marines engaged forces of the Barbary Coast at the Battle of Derna in Tripoli, marking the first recorded land battle by the United States on foreign soil.

  18. 1777year

    American Revolutionary War: British Army regulars defeated Patriot militias in the Battle of Ridgefield, galvanizing resistance in the Connecticut Colony.