Search: “models”

C-CoT: Counterfactual Chain-of-Thought with Vision-Language Models for Safe Autonomous Driving

arXiv:2605.10744v1 Announce Type: new Abstract: Safety-critical planning in complex environments, particularly at urban intersections, remains a fundamental challenge for autonomous driving. Existing

Research

DARE: Diffusion Language Model Activation Reuse for Efficient Inference

DGX agent

arXiv:2605.08134v1 Announce Type: cross Abstract: Diffusion Large Language Models (dLLMs) have emerged as a promising alternative to auto-regressive (AR) models, offering greater expressive capacity a

researcharxiv-cs-ai

12 May 2026

Do multimodal models imagine electric sheep?

arXiv:2605.09693v1 Announce Type: cross Abstract: Yes. We find that large multimodal models develop mental imagery when solving spatial puzzles, and they do imagine sheep when solving sheep puzzles. W

Agents

DriveFuture: Future-Aware Latent World Models for Autonomous Driving

arXiv:2605.09701v1 Announce Type: new Abstract: Existing latent world models for autonomous driving have opened a promising path toward future-aware driving intelligence. However, they typically treat

agentsarxiv-cs-cv

Local Ai

Dystruct: Dynamically Structured Diffusion Language Model Decoding via Bayesian Inference

arXiv:2605.09820v1 Announce Type: new Abstract: Diffusion language models (DLMs) have recently emerged as a promising alternative to autoregressive models, primarily due to their ability to enable par

local-aiarxiv-cs-lg

Edit-Based Refinement for Parallel Masked Diffusion Language Models

arXiv:2605.09603v1 Announce Type: new Abstract: Masked diffusion language models enable parallel token generation and offer improved decoding efficiency over autoregressive models. However, their perf

researcharxiv-cs-cl

Exploitation Without Deception: Dark Triad Feature Steering Reveals Separable Antisocial Circuits in Language Models

arXiv:2605.09773v1 Announce Type: cross Abstract: We use sparse autoencoder (SAE) feature steering to amplify Dark Triad personality traits (Machiavellianism, narcissism, and psychopathy) in Llama-3.3

Failing Forward: Adaptive Failure-Informed Learning for Vision-Language-Action Models

model-releasesarxiv-cs-ro

arXiv:2605.08434v1 Announce Type: new Abstract: Vision-language-action (VLA) models provide a promising paradigm for scalable robotic manipulation, yet their reliance on success-only behavioral clonin

Local Ai

FedGMI: Generative Model-Driven Federated Learning for Probabilistic Mixture Inference

arXiv:2605.08760v1 Announce Type: new Abstract: Federated Learning (FL) facilitates collaborative model training across decentralized clients while preserving data privacy by avoiding raw data exchang

local-aiarxiv-cs-lg

Filtering Memorization from Parameter-Space in Diffusion Models

arXiv:2605.10439v1 Announce Type: new Abstract: Low-Rank Adaptation (LoRA) has become a widely used mechanism for customizing diffusion models, enabling users to inject new visual concepts or styles t

Grounding the Score: Explicit Visual Premise Verification for Reliable Vision-Language Process Reward Models

arXiv:2603.16253v2 Announce Type: replace-cross Abstract: Vision-language process reward models (VL-PRMs) are increasingly used to score intermediate reasoning steps and rerank candidates under test-t

High-Entropy Tokens as Multimodal Failure Points in Vision-Language Models

arXiv:2512.21815v2 Announce Type: replace Abstract: Vision-language models (VLMs) achieve remarkable performance but remain vulnerable to adversarial attacks. Entropy, as a measure of model uncertaint

researcharxiv-cs-cv

Illusion-Aware Visual Preprocessing and Anti-Illusion Prompting for Classic Illusion Understanding in Vision-Language Models

arXiv:2605.08841v1 Announce Type: new Abstract: Vision-Language Models (VLMs) exhibit systematic bias toward visual illusions, recalling memorized facts rather than perceiving actual visual difference

Learngene Search Across Multiple Datasets for Building Variable-Sized Models

arXiv:2605.08209v1 Announce Type: new Abstract: Deep learning methods are widely used under diverse resource constraints, resulting in models of varying sizes, such as the Vision Transformer (ViT) ser

researcharxiv-cs-lg

Less Diverse, Less Safe: The Indirect But Pervasive Risk of Test-Time Scaling in Large Language Models

arXiv:2510.08592v3 Announce Type: replace-cross Abstract: Test-Time Scaling (TTS) improves LLM reasoning by exploring multiple candidate responses and then operating over this set to find the best out

LiteParse is the best open-source, model-free document parser for AI agents. Run it over over 50+ document types, and it will parse dense pa…

model-releasesjerry-liu--x

LiteParse is the best open-source, model-free document parser for AI agents. Run it over over 50+ document types, and it will parse dense pages with complex text layouts and tables, and it will extrac

LLiMba: Sardinian on a Single GPU -- Adapting a 3B Language Model to a Vanishing Romance Language

model-releasesarxiv-cs-cl

arXiv:2605.09015v1 Announce Type: new Abstract: Sardinian, a Romance language with roughly one million speakers, has minimal presence in modern NLP. Commercial services do not support it, and current

Med-StepBench: A Hierarchical Reasoning Framework for Evaluating Hallucinations in Medical Vision-Language Models

arXiv:2605.10002v1 Announce Type: new Abstract: Large vision-language models (VLMs) demonstrate strong performance in medical image understanding, but frequently generate clinically plausible yet inco

Meow-Omni 1: A Multimodal Large Language Model for Feline Ethology

model-releasesarxiv-cs-cl

arXiv:2605.09152v1 Announce Type: new Abstract: Deciphering animal intent is a fundamental challenge in computational ethology, largely because of semantic aliasing, the phenomenon where identical ext

On the Generation and Mitigation of Harmful Geometry in Image-to-3D Models

arXiv:2605.09606v1 Announce Type: cross Abstract: Recent advances in image-to-3D models have significantly improved the fidelity and accessibility of 3D content creation. Such a powerful reconstructio

safetyarxiv-cs-cv

Pretraining large language models with MXFP4

arXiv:2605.09825v1 Announce Type: cross Abstract: Why does full-pipeline FP4 training of large language models often diverge, even when forward activations and activation gradients remain stable? We a

SAID: Safety-Aware Intent Defense via Prefix Probing for Large Language Models

arXiv:2510.20129v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) remain vulnerable to jailbreak attacks, where adversarially crafted prompts induce policy-violating responses des

safetyarxiv-cs-ai

SLayerGen: a Crystal Generative Model for all Space and Layer Groups

arXiv:2605.08262v1 Announce Type: cross Abstract: Crystal generative models have shown rapid progress for accelerating the discovery of bulk, periodic materials. However, many material systems such as

SlimQwen: Exploring the Pruning and Distillation in Large MoE Model Pre-training

arXiv:2605.08738v1 Announce Type: cross Abstract: Structured pruning and knowledge distillation (KD) are typical techniques for compressing large language models, but it remains unclear how they shoul

This NVIDIA remains the strongest platform for large-model inference at scale. Prefill/decode disaggregation, Blackwell-native quantization,…

model-releasesperplexity--x

This NVIDIA remains the strongest platform for large-model inference at scale. Prefill/decode disaggregation, Blackwell-native quantization, custom kernels, and rack-scale NVLink turn GB200 into faste

Towards a Large Language-Vision Question Answering Model for MSTAR Automatic Target Recognition

arXiv:2605.10772v1 Announce Type: cross Abstract: Large language-vision models (LLVM), such as OpenAI's ChatGPT and GPT-4, have gained prominence as powerful tools for analyzing text and imagery. The

Towards Backdoor-Based Ownership Verification for Vision-Language-Action Models

arXiv:2605.09005v1 Announce Type: cross Abstract: Vision-Language-Action models (VLAs) support generalist robotic control by enabling end-to-end decision policies directly from multi-modal inputs. As

Towards Compact Sign Language Translation: Frame Rate and Model Size Trade-offs

model-releasesarxiv-cs-cl

arXiv:2605.09554v1 Announce Type: new Abstract: Sign Language Translation (SLT) converts sign language videos into spoken-language text, bridging communication between Deaf and hearing communities. Cu

Tracking the Truth: Object-Centric Spatio-Temporal Monitoring for Video Large Language Models

arXiv:2605.08974v1 Announce Type: cross Abstract: While multimodal large language models (MLLMs) have advanced video understanding, they remain highly prone to hallucinations in dynamic scenes. We arg

Uncovering Intra-expert Activation Sparsity for Efficient Mixture-of-Expert Model Execution

arXiv:2605.08575v1 Announce Type: cross Abstract: Mixture of Experts (MoE) architecture has become the standard for state-of-the-art large language models, owing to its computational efficiency throug

Understanding Asynchronous Inference Methods for Vision-Language-Action Models

arXiv:2605.08168v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models offer a promising path to generalist robot control, but their inference latency causes observation staleness when

Unsupervised Process Reward Models

arXiv:2605.10158v1 Announce Type: new Abstract: Process Reward Models (PRMs) are a powerful mechanism for steering large language model reasoning by providing fine-grained, step-level supervision. How

safetyarxiv-cs-lg

Amortized Multi-Objective Optimization Across Tasks with Generative Solution Modeling

model-releasesarxiv-cs-lg

arXiv:2511.09598v5 Announce Type: replace Abstract: Many real-world applications require solving families of expensive multi-objective optimization problems~(EMOPs) under varying operational condition

BGM-IV: an AI-powered Bayesian generative modeling approach for instrumental variable analysis

arXiv:2605.07029v1 Announce Type: cross Abstract: Instrumental-variable (IV) regression enables causal estimation under endogeneity, but modern IV problems often involve nonlinear structural effects a

DT-PBO: an Interpretable Tree-based Surrogate Model for Preferential Bayesian Optimization

arXiv:2512.14263v2 Announce Type: replace-cross Abstract: Preferential Bayesian Optimization (PBO) aims to find a decision-maker's most preferred solution in as few pairwise comparisons as possible. E

Efficient Data Selection for Multimodal Models via Incremental Optimization Utility

arXiv:2605.07488v1 Announce Type: new Abstract: The scaling of Large Multimodal Models (LMMs) is constrained by the quality-quantity trade-off inherent in synthetic data. Previous approaches, such as

Flow-OPD: On-Policy Distillation for Flow Matching Models

arXiv:2605.08063v1 Announce Type: cross Abstract: Existing Flow Matching (FM) text-to-image models suffer from two critical bottlenecks under multi-task alignment: the reward sparsity induced by scala

safetyarxiv-cs-ai

Frequency-Aware Model Parameter Explorer: A new attribution method for improving explainability

arXiv:2510.03245v2 Announce Type: replace-cross Abstract: State-of-the-art attribution methods rely on adversarial sample generation that applies an all-pass filter across the frequency spectrum, disc

Introducing Daybreak: frontier AI for cyber defenders. Daybreak brings together the most capable OpenAI models, Codex, and our security part…

Introducing Daybreak: frontier AI for cyber defenders. Daybreak brings together the most capable OpenAI models, Codex, and our security partners to accelerate cyber defense and continuously secure sof

model-releasesopenai--x

Applications

It is one reason why I think the push for smaller, local models is more complicated than people think. If you want good answers, especially …

applicationsethan-mollick--x

It is one reason why I think the push for smaller, local models is more complicated than people think. If you want good answers, especially good answers to unexpected problems, frontier models will ge

MicroBi-ConvLSTM: An Ultra-Lightweight Efficient Model for Human Activity Recognition on Resource Constrained Devices

arXiv:2602.06523v2 Announce Type: replace Abstract: Human Activity Recognition (HAR) on resource constrained wearables requires models that balance accuracy against strict memory and computational bud

NoiseGate: Learning Per-Latent Timestep Schedules as Information Gating in World Action Models

arXiv:2605.07794v1 Announce Type: new Abstract: World Action Models (WAMs) are an emerging family of policies that tie robot action generation to future-observation modeling. In this work, we focus on

safetyarxiv-cs-ro

Normalizing Trajectory Models

arXiv:2605.08078v1 Announce Type: new Abstract: Diffusion-based models decompose sampling into many small Gaussian denoising steps -- an assumption that breaks down when generation is compressed to a

researcharxiv-cs-cv

Applications

Our research, as well as that of other researchers, shows better prompting techniques help a lot, but model training is still a huge limitin…

applicationsethan-mollick--x

Ethan Mollick discusses research findings showing that while improved prompting techniques provide significant benefits for AI model performance, the underlying model training remains the primary limi

Outlier Smoothing with Closed-Form Rotations for W4A4 Large Language Model Quantization

model-releasesarxiv-cs-lg

arXiv:2511.22316v2 Announce Type: replace Abstract: Large Language Models (LLMs) quantization facilitates deploying LLMs in resource-limited settings, but existing methods that combine incompatible gr

PathPainter: Transferring the Generalization Ability of Image Generation Models to Embodied Navigation

model-releasesarxiv-cs-ro

arXiv:2605.07496v1 Announce Type: new Abstract: Bird's-eye-view (BEV) images have been widely demonstrated to provide valuable prior information for navigation. Given the global information provided b

PolarVLM: Bridging the Semantic-Physical Gap in Vision-Language Models

arXiv:2605.07574v1 Announce Type: new Abstract: Mainstream vision-language models (VLMs) fundamentally struggle with severe optical ambiguities, such as reflections and transparent objects, due to the

ProteinJEPA: Latent prediction complements protein language models

arXiv:2605.07554v1 Announce Type: cross Abstract: Protein language models are trained primarily with masked language modeling (MLM), which predicts amino-acid identities at masked positions. We ask wh