tutorials

Do AI models really learn physics, or just learn what physics looks like? Flatiron's @_helenqu, with @PolymathicAI & CDS researchers including @ylecun, finds that models predicting in latent space rec

arXiv:2605.11527v1 Announce Type: new Abstract: Diffusion models are the leading approach for tabular data synthesis and are increasingly used to share sensitive records. Whether they actually protect

In this post, we demonstrate how to build a secure, complete LLM fine-tuning workflow that integrates Unity Catalog with Amazon SageMaker AI using Amazon EMR Serverless for preprocessing. The solution

arXiv:2602.00297v2 Announce Type: replace Abstract: Deep learning has achieved strong performance in Time Series Forecasting (TSF). However, we identify a critical representation paradox, termed Laten

arXiv:2511.15572v2 Announce Type: replace Abstract: Feature-map knowledge distillation (KD) transfers internal representations well between comparably sized Vision Transformers (ViTs), but it often fa

Introducing GridSFM, a small foundation model that can predict AC optimal power flow in milliseconds, boosting efficiency and unlocking cost savings. Learn how GridSFM gives grid operators direct visi

arXiv:2605.12369v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models aim for general robot learning by aligning action as a modality within powerful Vision-Language Models (VLMs). Exist

How to Share Your ComfyUI App Mode Workflow on Comfy Cloud Once you’ve built an app, you can share it with a single URL. The URL encodes everything a recipient needs to open and run your app: the work

i provided a clear road map in the essay Next Decade in AI in arxiv in 2020. not one person in the field has given a serious critique of it. if you are not aware of it, it is not my fault; i have poin

arXiv:2605.12013v1 Announce Type: new Abstract: Pixel diffusion models have recently regained attention for visual generation. However, training advanced pixel-space models from scratch demands prohib

arXiv:2510.16127v2 Announce Type: replace-cross Abstract: The ratio of two probability density functions is a fundamental quantity that appears in many areas of statistics and machine learning, includ

arXiv:2605.11749v1 Announce Type: new Abstract: Weakly supervised graph anomaly detection aims to unveil unusual graph instances, e.g., nodes, whose behaviors significantly differ from normal ones, gi

arXiv:2603.23679v2 Announce Type: replace Abstract: Agriculture remains a cornerstone of global health and economic sustainability, yet labor-intensive tasks such as harvesting high-value crops contin

// δ-mem: Efficient Online Memory for LLMs // One of the more elegant memory mechanisms I've seen this month. Most long-term memory work either inflates context or retrains the model. This paper shows

Maybe I'm just punch-drunk in my third week attending Musk v. Altman, but I have become very, very fond of Microsoft during the course of this trial. They don't want to be here any more than I do. The

arXiv:2605.12242v1 Announce Type: new Abstract: Automatic Speech Recognition (ASR) transcripts often contain disfluencies, such as fillers, repetitions, and false starts, which reduce readability and

Mother's Day may have passed, but we're not done showcasing mothers who build on Replit. When Noni started a family, she put her dream of building apps on hold. A busy home, a busy classroom, and no t

arXiv:2509.21711v2 Announce Type: replace-cross Abstract: As data collection and simulation capabilities advance, multi-modal learning, the task of learning from multiple modalities and sources of dat

arXiv:2302.12039v2 Announce Type: replace Abstract: We summarize the current state of the field of NLP & Law with a specific focus on recent technical and substantive developments. To support our anal

arXiv:2605.12168v1 Announce Type: new Abstract: Artificial intelligence systems typically rely on large, centrally collected datasets, a premise that does not hold in many real-world domains such as h

arXiv:2605.11742v1 Announce Type: new Abstract: Online Continual Learning (OCL) aims to learn from endless nonext{-}stationary data streams, yet most existing methods assume a flat label space and ove

OpenAI details its response to the TanStack “Mini Shai-Hulud” supply chain attack, outlines protections taken to secure systems and signing certificates, and explains why macOS users must update OpenA

arXiv:2605.11266v1 Announce Type: new Abstract: Recent advances in Gaussian Splatting have enabled fast, high-fidelity 3D scene generation, yet these methods remain purely visual and lack an understan

arXiv:2508.10036v2 Announce Type: replace Abstract: Large Language Models (LLMs) show remarkable potential for few-shot information extraction (IE), yet their performance is highly sensitive to the ch

arXiv:2605.12494v1 Announce Type: new Abstract: Surface reconstruction with differentiable rendering has achieved impressive performance in recent years, yet the pervasive photometric ambiguities have

arXiv:2605.11547v1 Announce Type: new Abstract: Flow matching models generate samples by numerically integrating a learned velocity field, with each integration step requiring a neural network evaluat

arXiv:2504.10428v2 Announce Type: replace-cross Abstract: Binary classification from positive-only samples is a variant of PAC learning where the learner receives i.i.d. positive samples and aims to l

arXiv:2602.05352v2 Announce Type: replace Abstract: Modern neural networks have shown promise for solving partial differential equations over surfaces, often by discretizing the surface as a mesh and

arXiv:2506.09044v2 Announce Type: replace Abstract: Machine Learning systems are increasingly deployed in decision-making settings that shape user behavior and, in turn, the data on which future decis

arXiv:2503.18760v2 Announce Type: replace Abstract: A key consideration when training an LLM is whether the target language is more or less resourced, for example English compared to Welsh, or Python

arXiv:2602.19770v2 Announce Type: replace Abstract: Explainable artificial intelligence has emerged as a promising field of research to address reliability concerns in artificial intelligence. Despite

TST has two moving parts that can be ablated independently: the input-side averaging and the output-side multi-hot cross-entropy. Both, on their own, outperform the matched-FLOPs baseline. Their effec

arXiv:2605.12088v1 Announce Type: new Abstract: Multi-reference image generation aims to synthesize images from textual instructions while faithfully preserving subject identities from multiple refere

arXiv:2605.12281v1 Announce Type: new Abstract: What makes a word difficult to learn, and how does the difficulty depend on the learner's native language? We computationally model vocabulary difficult

arXiv:2605.11008v1 Announce Type: new Abstract: While invariant architectures are standard for processing symmetric data, there is growing interest in achieving invariance by applying group averaging

arXiv:2605.08432v1 Announce Type: cross Abstract: Calibration measures whether a model's predicted confidence aligns with its empirical accuracy, and is central to the reliable deployment of large lan

arXiv:2605.10654v1 Announce Type: cross Abstract: We consider the active learning problem where the goal is to learn an unknown function with low prediction error under an unknown Boltzmann distributi

arXiv:2605.09995v1 Announce Type: new Abstract: Post-training (via supervised fine-tuning) improves instruction-following, but often induces semantic mode collapse by biasing models toward low-entropy

arXiv:2605.09492v1 Announce Type: cross Abstract: Large language models (LLMs) often suffer from hallucinations due to error accumulation in autoregressive decoding, where suboptimal early token choic

arXiv:2605.08679v1 Announce Type: cross Abstract: Graph neural networks (GNNs) aim to learn well-trained representations in a lower-dimension space for downstream tasks while preserving the topologica

arXiv:2605.09992v1 Announce Type: cross Abstract: Speculative decoding accelerates LLM inference by drafting future tokens with a small model, but drafter models degrade sharply under template perturb

In this post, we'll show you how our multi-document discovery feature solves this problem. It serves as an automated pre-processing step, analyzing unknown documents, clustering them by type, and gene

AutoScout24, a leading European online automotive marketplace, implemented AI-powered workflows to optimize and scale their engineering operations. The implementation likely leveraged OpenAI's technol

Silencio leveraged Vultr's full-stack infrastructure platform to optimize their operations, likely achieving improvements in performance, scalability, or cost efficiency through Vultr's cloud computin

arXiv:2605.10586v1 Announce Type: new Abstract: Learning a physical model from video data that can comprehend physical laws and predict the future trajectories of objects is a formidable challenge in

arXiv:2605.09079v1 Announce Type: new Abstract: Despite surpassing human performance across mathematics, coding, and other knowledge-intensive tasks, large language models (LLMs) continue to struggle