Search: “models”

Beyond Linear Superposition: Discovering Climate Features in AI Weather Models with KAN-SAE

arXiv:2605.17493v1 Announce Type: cross Abstract: Deep learning weather prediction models achieve remarkable predictive skill yet remain largely opaque: we know little about how they represent physica

Safety

Beyond the Final Actor: Modeling the Dual Roles of Creator and Editor for Fine-Grained LLM-Generated Text Detection

DGX agent

arXiv:2604.04932v3 Announce Type: replace Abstract: The misuse of large language models (LLMs) requires precise detection of synthetic text. Existing works mainly follow binary or ternary classificati

safetyarxiv-cs-cl

19 May 2026

Model Releases

Can These Views Be One Scene? Evaluating Multiview 3D Consistency when 3D Foundation Models Hallucinate

model-releasesarxiv-cs-cv

arXiv:2605.18754v1 Announce Type: new Abstract: Multiview 3D evaluation assumes that the images being scored are observations of one static 3D scene. This assumption can fail in NVS and sparse-view re

Concise and Logically Consistent Conformal Sets for Neuro-Symbolic Concept-Based Models

arXiv:2605.18202v1 Announce Type: cross Abstract: Neuro-Symbolic Concept-based Models (NeSy-CBMs) are a family of architectures that integrate neural networks with symbolic reasoning for enhanced reli

CyberCorrect: A Cybernetic Framework for Closed-Loop Self-Correction in Large Language Models

arXiv:2605.17305v1 Announce Type: new Abstract: Large language model (LLM) self-correction -- the ability to detect and fix errors in generated outputs -- remains largely ad hoc, relying on generic pr

Deep sequence models tend to memorize geometrically; it is unclear why

arXiv:2510.26745v3 Announce Type: replace-cross Abstract: Deep sequence models are said to store atomic facts predominantly in the form of associative memory: a brute-force lookup of co-occurring enti

Dimension-Free Convergence of Discrete Diffusion Models: Adjoint Equations Induce the Right Space

arXiv:2605.17232v1 Announce Type: new Abstract: Discrete diffusion has become a leading framework for generative modeling in various applications including language, vision, and biology. Existing conv

Distilling Tabular Foundation Models for Structured Health Data

arXiv:2605.18702v1 Announce Type: cross Abstract: Tabular foundation models (TFMs) achieve strong performance on health datasets, but their inference cost and infrastructure requirements limit practic

DreamEdit3D: Personalization of Multi-View Diffusion Models for 3D Editing

arXiv:2605.16990v1 Announce Type: new Abstract: While 2D diffusion models have achieved remarkable success in identity-preserving personalization, extending this capability to 3D assets remains a sign

Early Stopping Chain-of-thoughts in Large Language Models

arXiv:2509.14004v2 Announce Type: replace Abstract: Reasoning large language models (LLMs) have demonstrated superior capacities in solving complicated problems by generating long chain-of-thoughts (C

Entropy-Gradient Inversion: Moving Toward Internal Mechanism of Large Reasoning Models

arXiv:2605.17770v1 Announce Type: new Abstract: The advancement of Large Reasoning Models (LRMs) has catalyzed a paradigm shift from reactive ``fast thinking'' text generation to systematic, step-by-s

Applications

Exploring Lightweight Large Language Models for Court View Generation

arXiv:2605.16770v1 Announce Type: cross Abstract: Criminal Court View Generation (CVG) is a critical task in Legal Artificial Intelligence (Legal AI), involving the generation of court view based on c

applicationsarxiv-cs-ai

Exploring Trust Calibration in XAI - The Impact of Exposing Model Limitations to Lay Users

arXiv:2605.18036v1 Announce Type: cross Abstract: Trust calibration -- aligning user trust judgment with model capability -- is crucial for safe deployment of explainable AI (XAI), yet is often evalua

Forget Many, Forget Right: Scalable and Precise Concept Unlearning in Diffusion Models

arXiv:2601.06162v4 Announce Type: replace-cross Abstract: Text-to-image diffusion models have achieved remarkable progress, yet their use raises copyright and misuse concerns, prompting research into

GeoWorld: Geometric World Models

arXiv:2602.23058v2 Announce Type: replace Abstract: Energy-based predictive world models provide a powerful approach for multi-step visual planning by reasoning over latent energy landscapes rather th

researcharxiv-cs-cv

Global Prior Meets Local Consistency: Dual-Memory Augmented Vision-Language-Action Model for Efficient Robotic Manipulation

arXiv:2602.20200v2 Announce Type: replace-cross Abstract: Hierarchical Vision-Language-Action (VLA) models have rapidly become a dominant paradigm for robotic manipulation. It typically comprising a V

local-aiarxiv-cs-ai

HDTree: Generative Modeling of Cellular Hierarchies for Robust Lineage Inference

arXiv:2506.23287v3 Announce Type: replace Abstract: In single-cell research, tracing and analyzing high-throughput single-cell differentiation trajectories is crucial for understanding biological proc

Applications

HighSync: High-Quality Lip Synchronization via Latent Diffusion Models

arXiv:2605.16918v1 Announce Type: new Abstract: We present HighSync, an end-to-end diffusion-based framework for high-fidelity lip synchronization that generates photorealistic talking-face videos ali

applicationsarxiv-cs-cv

Tutorials

How to run evals for the model router

tutorialsmicrosoft-foundry

Walk through running quality, cost, and latency evaluations for the Foundry model router using an open-source GitHub repo designed for router-aware eval pipelines. The post How to run evals for the mo

I made a tool to use AgentRouter models in OpenCode

The search results focus on OpenCode + Ollama integration but don't specifically cover the AgentRouter tool. Let me provide a knowledge base entry based on what the title and context suggest: A tool d

local-air-ollama

Interpretable epistemic uncertainty decomposition in sequential generative models via polynomial chaos surrogates

arXiv:2510.21523v2 Announce Type: replace Abstract: Sequential generative models conditioned on uncertain rewards are central to AI-driven scientific discovery, yet the epistemic uncertainty they inhe

safetyarxiv-cs-lg

LatentUMM: Dual Latent Alignment for Unified Multimodal Models

arXiv:2605.17766v1 Announce Type: new Abstract: Unified multimodal models (UMMs) achieve strong performance in both understanding and generation by learning a shared latent space, yet they often exhib

Lipschitz-Guided Design of Interpolation Schedules in Generative Models

arXiv:2509.01629v3 Announce Type: replace-cross Abstract: We study the design of interpolation schedules in flow and diffusion-based generative models from both statistical and numerical perspectives.

Medical Context Distorts Decisions in Clinical Vision Language Models

arXiv:2605.17436v1 Announce Type: cross Abstract: Vision-language models (VLMs) are increasingly proposed for clinical decision support, yet their reliability in real-world scenarios that require inte

safetyarxiv-cs-cl

Applications

Mixing Times of Glauber Dynamics on Masked Language Models

arXiv:2605.16378v1 Announce Type: cross Abstract: Masked language models (MLMs) define local conditional distributions over tokens but do not, in general, correspond to any consistent joint distributi

applicationsarxiv-cs-ai

MV-Gate: Insider Threat Detection via Multi-View Behavioral Statistics and Semantic Modeling

arXiv:2605.17761v1 Announce Type: cross Abstract: Insider threats often reveal early anomalies through disruptions in behavioral statistics-such as altered recurrence patterns or short-versus long-ter

Naturalistic Computational Cognitive Science: Towards generalizable models and theories that capture the full range of natural behavior

arXiv:2502.20349v3 Announce Type: replace-cross Abstract: How can cognitive science build generalizable theories that span the full scope of natural situations and behaviors? We argue that progress in

NeuroRVQ: Multi-Scale Biosignal Tokenization for Generative Foundation Models

arXiv:2510.13068v4 Announce Type: replace-cross Abstract: Biosignals such as electroencephalography (EEG), electrocardiography (ECG), and electromyography (EMG) encode physiological activity across mu

Olivia: Harmonizing Time Series Foundation Models with Power Spectral Density

arXiv:2605.17340v1 Announce Type: new Abstract: Time series foundation models rely on large-scale pretraining over diverse datasets across domains, yet their heterogeneity in temporal patterns could h

Model Releases

POST: Prior-Observation Adversarial Learning of Spatio-Temporal Associations for Multivariate Time Series Anomaly Detection

model-releasesarxiv-cs-ai

arXiv:2605.18128v1 Announce Type: new Abstract: Existing Multivariate Time Series Anomaly Detection (MTSAD) frameworks increasingly rely on integrating Graph Neural Networks (GNNs) with sequence model

Public-Decay Homomorphic State Space Models for Private Sequence Inference

arXiv:2605.16647v1 Announce Type: cross Abstract: Fully homomorphic encryption (FHE) changes sequence-model design because rotations, encrypted products, ciphertext materialization, multiplicative dep

local-aiarxiv-cs-lg

RAVE: Re-Allocating Visual Attention in Large Multimodal Models

arXiv:2605.18359v1 Announce Type: new Abstract: Large multimodal models (LMMs) inherit the self-attention mechanism of pretrained language backbones, yet standard attention can exhibit suboptimal allo

Response-free item difficulty modelling for multiple-choice items with fine-tuned transformers: Component-wise representation and multi-task learning

arXiv:2605.16991v1 Announce Type: cross Abstract: Response-free item difficulty modelling promises to reduce reliance on response-based calibration but is intrinsically difficult on reading-comprehens

Running the same CNA search on base models (before instruction tuning) yields a structurally similar set of neurons, but ablating them produ…

Running the same CNA search on base models (before instruction tuning) yields a structurally similar set of neurons, but ablating them produces almost no behavioral change. We read this as evidence th

safetynous-research--x

Self-Distillation is Optimal Among Spectral Shrinkage Estimators in Spiked Covariance Models

arXiv:2605.17778v1 Announce Type: cross Abstract: Self-distillation has emerged as a promising technique for improving model performance in modern machine learning systems. We develop the statistical

Semantic Generative Tuning for Unified Multimodal Models

arXiv:2605.18714v1 Announce Type: cross Abstract: Unified multimodal models (UMMs) strive to consolidate visual understanding and visual generation within a single architecture. However, prevailing tr

Simple Approximation and Derivative Free Inference-Time Scaling for Diffusion Models via Sequential Monte Carlo on Path Measures

arXiv:2605.17850v1 Announce Type: cross Abstract: iffusion-based generative models increasingly rely on inference-time guidance, adding a drift term or reweighting mixture of experts, to improve sampl

Single-Sample Black-Box Membership Inference Attack against Vision-Language Models via Cross-modal Semantic Alignment

arXiv:2605.17341v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have achieved remarkable success, yet their reliance on massive datasets and unintended memorization of training data ra

Tutorials

Sparse Autoencoders are Topic Models

arXiv:2511.16309v2 Announce Type: replace Abstract: Sparse autoencoders (SAEs) are used to analyze embeddings, but their role and practical value are debated. We propose a new perspective on SAEs by d

tutorialsarxiv-cs-cv

Speak Your Mind: The Speech Continuation Task as a Probe of Voice-Based Model Bias

arXiv:2509.22061v2 Announce Type: replace-cross Abstract: Speech Continuation (SC) is the task of generating a coherent extension of a spoken prompt while preserving both semantic context and speaker

safetyarxiv-cs-cl

Spherical Steering: Geometry-Aware Activation Rotation for Language Models

arXiv:2602.08169v2 Announce Type: replace-cross Abstract: Inference-time steering offers a promising way to control language models (LMs) without retraining. However, standard approaches typically rel

Stop When Reasoning Converges: Semantic-Preserving Early Exit for Reasoning Models

arXiv:2605.17672v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) achieve strong performance by generating long chains of thought (CoT), but often overthink, continuing to reason after a s

safetyarxiv-cs-cl

Model Releases

Sustainability via LLM Right-sizing

model-releasesarxiv-cs-ai

arXiv:2504.13217v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have become increasingly embedded in organizational workflows. This has raised concerns over their energy consump

The Expert Strikes Back: Interpreting Mixture-of-Experts Language Models at Expert Level

arXiv:2604.02178v2 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) architectures have become the dominant choice for scaling Large Language Models (LLMs), activating only a subset of p

The Frequency Confound in Language-Model Surprisal and Metaphor Novelty

arXiv:2605.06506v2 Announce Type: replace Abstract: Language-model (LM) surprisal is widely used as a proxy for contextual predictability and has been reported to correlate with metaphor novelty judgm

The Point of No Return: Counterfactual Localization of Deceptive Commitment in Language-Model Reasoning

arXiv:2605.17113v1 Announce Type: cross Abstract: Existing deception datasets label completed outputs as honest or deceptive, treating deception as a property of the final response rather than a funct

local-aiarxiv-cs-ai

The Unlearnability Phenomenon in RLVR for Language Models

arXiv:2605.16787v1 Announce Type: cross Abstract: Reinforcement Learning with Verifiable Reward (RLVR) has proven effective in improving Large Language Model's (LLM) reasoning ability. However, the le

Hardware

TierCheck: Tiered Checkpointing for Fault Tolerance in Large Language Model Training

arXiv:2605.17821v1 Announce Type: cross Abstract: Large Language Model (LLM) training is frequently interrupted by a heterogeneous spectrum of failures, from common GPU crashes to catastrophic cluster

hardwarearxiv-cs-ai