Persona-Model Collapse in Emergent Misalignment
arXiv:2605.12850v1 Announce Type: cross Abstract: Fine-tuning large language models on narrow data with harmful content produces broadly misaligned behavior on unrelated prompts, a phenomenon known as
Knowledge catalogue
arXiv:2605.12850v1 Announce Type: cross Abstract: Fine-tuning large language models on narrow data with harmful content produces broadly misaligned behavior on unrelated prompts, a phenomenon known as
arXiv:2601.09636v2 Announce Type: replace Abstract: While GUI agents have shown strong performance under explicit and completion instructions, real-world deployment requires aligning with users' more
arXiv:2605.12541v1 Announce Type: cross Abstract: Electrocardiography (ECG) is the clinical standard for cardiac assessment but requires dedicated hardware that does not scale to daily-life monitoring
arXiv:2605.13370v1 Announce Type: new Abstract: For over a decade, explicit memory architectures like the Neural Turing Machine have remained theoretically appealing yet practically intractable for la
arXiv:2605.13583v1 Announce Type: new Abstract: Recent advances have demonstrated that coded aperture snapshot spectral imaging (CASSI) systems show great potential for capturing 3D hyperspectral imag
arXiv:2605.13493v1 Announce Type: new Abstract: Can general-purpose image editors predict physical maps from a single RGB image? General-purpose image editors differ from standard task-specific dense-
arXiv:2605.13268v1 Announce Type: cross Abstract: Product formulas for Trotter Suzuki simulation remain a practical route to Hamiltonian evolution on noisy intermediate scale quantum (NISQ) hardware,
arXiv:2602.23089v2 Announce Type: replace Abstract: The Bayesian update step poses significant computational challenges in high-dimensional nonlinear estimation. While log-homotopy particle flow filte
arXiv:2605.12803v1 Announce Type: new Abstract: Detecting concept drift in high-speed data streams remains challenging, particularly when models must operate on unlabeled data and avoid false alarms c
arXiv:2605.12653v1 Announce Type: cross Abstract: Reinforcement learning agents for portfolio management are typically trained and deployed as static policies, with no mechanism for using price foreca
arXiv:2506.19037v4 Announce Type: replace-cross Abstract: Masked diffusion language models (MDLMs) promise fast, non-autoregressive text generation, yet existing samplers, which pick tokens to unmask
Please don't use the offensive word 'rape.' It is Undocumented Love Making as I explain in Suicidal Empathy. Elon Musk: 'These lovely small towns in England, Scotland, Ireland, are living their lives
arXiv:2602.02791v2 Announce Type: replace-cross Abstract: We study supervised multiclass classification for diffusion processes, where each class is characterized by a distinct drift function and traj
arXiv:2605.13692v1 Announce Type: new Abstract: Many online decision problems over combinatorial actions are addressed via convex relaxations, leading to online convex optimization with piecewise line
arXiv:2601.18608v3 Announce Type: replace Abstract: Shapley values have emerged as a central game-theoretic tool in explainable AI (XAI). However, computing Shapley values exactly requires 2^d game ev
arXiv:2605.12648v1 Announce Type: new Abstract: We establish the first population risk bounds for Kolmogorov-Arnold Networks (KANs) trained by mini-batch SGD with gradient clipping, covering non-priva
arXiv:2605.12966v1 Announce Type: new Abstract: Is monolithic scaling the only path to AGI? This paper challenges the dogma that purely scaling a single model is sufficient to achieve Artificial Gener
arXiv:2605.13579v1 Announce Type: new Abstract: Assistive agents for Blind and Visually Impaired (BVI) users require accessibility alignment as a first-class design objective. Despite rapid progress i
arXiv:2507.22095v5 Announce Type: replace-cross Abstract: We consider fully connected and feedforward deep neural networks with dependent and possibly heavy-tailed weights, as introduced in [26], to a
arXiv:2508.14950v2 Announce Type: replace-cross Abstract: 4D Flow Magnetic Resonance Imaging (4D Flow MRI) enables non-invasive quantification of blood flow and hemodynamic parameters. However, its cl
arXiv:2605.13081v1 Announce Type: new Abstract: Missing modalities are prevalent in real-world Alzheimer's disease (AD) assessment and pose a significant challenge to multimodal learning, particularly
arXiv:2602.06104v2 Announce Type: replace Abstract: Many engineering and scientific workflows rely on expensive black-box evaluations, requiring sequential decisions that must both improve task perfor
arXiv:2605.12505v1 Announce Type: cross Abstract: Autonomous AI systems generate responsibility gaps: consequential actions that cannot be satisfactorily attributed to developers, operators, or users
arXiv:2603.02337v2 Announce Type: replace-cross Abstract: Flow matching (FM) learns vector fields by regressing stochastic velocity targets along intermediate distributions p_t. We identify a geometri
arXiv:2601.21033v2 Announce Type: replace Abstract: Diffusion models cannot enforce hard constraints, yet applications in the physical sciences demand exact satisfaction of conservation laws, boundary
arXiv:2605.12759v1 Announce Type: new Abstract: The Lightning Network (LN) is a second-layer protocol for Bitcoin designed to enable fast and cost-efficient off-chain transactions. Channels in the LN
arXiv:2605.12855v1 Announce Type: new Abstract: Clinical trial studies indicate benefit of watch-and-wait (WW) surveillance for patients with rectal cancer showing a complete or near clinical response
arXiv:2605.13396v1 Announce Type: new Abstract: Face Image Quality Assessment (FIQA) evaluates the utility of a face image for automated face recognition (FR) systems. In this work, we propose PreFIQs
arXiv:2602.02001v2 Announce Type: replace-cross Abstract: Quantization Error Reconstruction (QER) reduces accuracy loss in Post-Training Quantization (PTQ) by approximating weights as mathbf{W} approx
arXiv:2605.13436v1 Announce Type: cross Abstract: Subword regularization methods such as BPE dropout are typically applied only during fine-tuning, while pretraining is usually done with deterministic
arXiv:2605.12504v1 Announce Type: cross Abstract: We develop conjectures and theorems expressing the idea that the prime sequence exhibits computational irreducibility in the transition from one prime
arXiv:2605.12851v1 Announce Type: cross Abstract: Automated analysis of peripheral blood smears for Acute Lymphoblastic Leukemia (ALL) is hindered by low contrast and substantial variability in cytopl
arXiv:2605.13027v1 Announce Type: new Abstract: Text image super-resolution (Text-SR) requires more than visually plausible detail synthesis: slight errors in stroke topology may alter character ident
arXiv:2512.08411v2 Announce Type: replace Abstract: Model-based planning in robotic domains is challenged by the hybrid nature of physical dynamics, where continuous motion is punctuated by discrete e
arXiv:2510.13385v3 Announce Type: replace Abstract: Although both data availability and the demand for accurate forecasts are increasing, collaboration between stakeholders is often constrained by dat
arXiv:2605.13339v1 Announce Type: cross Abstract: Large language models (LLMs) can be said to have preferences: they reliably pick certain tasks and outputs over others, and preferences shaped by post
arXiv:2603.20334v2 Announce Type: replace-cross Abstract: In high-complexity abstract reasoning, a system must infer a latent rule from a few examples or structured observations and apply it to unseen
arXiv:2605.12664v1 Announce Type: cross Abstract: Bilateral trade models the task of intermediating between two strategic agents, a seller and a buyer, who wish to trade a good. We study this problem
arXiv:2510.00417v2 Announce Type: replace-cross Abstract: An algorithm is proposed, analyzed, and tested for solving continuous nonlinear-equality-constrained optimization problems where the objective
arXiv:2605.12835v1 Announce Type: new Abstract: Large language models can extract local causal claims from text, but those claims become more useful when organized as persistent, navigable world model
Protected Source Maps is a Vercel feature that enables developers to securely ship browser source maps to production without exposing them publicly. This allows for proper error tracking and debugging
arXiv:2602.12026v2 Announce Type: replace Abstract: Protein language models (pLMs) have emerged as powerful predictors of protein structure and function. However, the computational circuits underlying
arXiv:2605.12981v1 Announce Type: cross Abstract: Automated program synthesis has reduced the cost of producing candidate implementations, but it introduces a harder governance problem: determining wh
arXiv:2605.13810v1 Announce Type: new Abstract: Vector quantization via random projection followed by scalar quantization is a fundamental primitive in machine learning, with applications ranging from
arXiv:2602.06239v2 Announce Type: replace Abstract: We introduce PEPO (Pessimistic Ensemble based Preference Optimization), a single-step Direct Preference Optimization (DPO)-like algorithm to mitigat
arXiv:2605.13278v1 Announce Type: cross Abstract: Score-based diffusion models demonstrate superior performance in generative tasks but encounter fundamental bottlenecks in inverse problems due to the
PricewaterhouseCoopers LLP and Anthropic PBC today announced a major expansion of their alliance, as the consulting giant is set to roll out Claude Code and Claude Cowork across its workforce and trai
arXiv:2605.13111v1 Announce Type: new Abstract: Autoregressive video generation enables streaming and open-ended long video synthesis, but still suffers from long-term degradation caused by accumulate
arXiv:2605.12567v1 Announce Type: cross Abstract: The inherent electronic and speckle noise complicates clinical interpretation of ultrasound images. Conventional denoising methods rely on explicit no
arXiv:2605.13435v1 Announce Type: cross Abstract: There is growing interest in utilizing flow-based models as decision-making policies in reinforcement learning due to their high expressive capacity.
arXiv:2605.13833v1 Announce Type: cross Abstract: Modeling long-range dependencies in sequential data remains a central challenge in machine learning. Transformers address this challenge through atten
arXiv:2512.16960v2 Announce Type: replace Abstract: Quantum-inspired machine learning (QiML) employs mathematical principles from quantum theory, such as Hilbert-space representations and quantum stat
arXiv:2605.12869v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in a wide range of applications, yet remain vulnerable to adversarial jailbreak attacks that ci
arXiv:2605.12831v1 Announce Type: new Abstract: Inverse reinforcement learning (IRL), which infers reward functions from demonstrations, is a valuable tool for modeling and understanding decision-maki
arXiv:2605.13830v1 Announce Type: new Abstract: Decision tree ensembles (DTE) are a popular model for a wide range of AI classification tasks, used in multiple safety critical domains, and hence verif
arXiv:2510.03992v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly deployed in agentic systems, where a fundamental task is mapping user intents to relevant extern
arXiv:2605.13369v1 Announce Type: cross Abstract: Large language models (LLMs) are typically deployed with fixed parameters, and their performance is often improved by allocating more computation at i
arXiv:2511.17855v2 Announce Type: replace Abstract: Robots must learn from both what people do and what they say, but either modality alone is often incomplete: physical corrections are grounded but a
[...] On the interesting side is how fungible programming languages are nowadays. Programming languages used to be LOCK IN, and they're increasingly not so. You think the Bun rewrite in Rust is good f
arXiv:2605.13565v1 Announce Type: new Abstract: We present Qwen-Image-VAE-2.0, a suite of high-compression Variational Autoencoders (VAEs) that achieve significant advances in both reconstruction fide