All entries

arXiv:2607.22723v1 Announce Type: new Abstract: Visual information extraction (VIE) from visually rich documents remains challenging due to high layout variability and real-world impairments. Existing

applicationsarxiv-cs-cv

Visual Token Compression Enhances Robustness of MLLMs

DGX agent

arXiv:2607.22716v1 Announce Type: new Abstract: In this paper, we show for the first time that visual token pruning enhances the robustness of Multimodal Large Language Models (MLLMs), mitigating vuln

safetyarxiv-cs-cv

28 Jul 2026

VL-LN Bench: Towards Long-horizon Goal-oriented Navigation with Active Dialogs

model-releasesarxiv-cs-ro

arXiv:2512.22342v5 Announce Type: replace Abstract: In most existing embodied navigation tasks, instructions are well-defined and unambiguous, such as instruction following and object searching. Under

VLASH: Real-Time VLAs via Future-State-Aware Asynchronous Inference

arXiv:2512.01031v2 Announce Type: replace-cross Abstract: Vision-Language-Action models (VLAs) are becoming increasingly capable across diverse robotic tasks. However, these models are typically deplo

VlogReward: Learning Multi-Dimensional Evaluation for Vlog Editing

model-releasesarxiv-cs-ai

arXiv:2607.22632v1 Announce Type: new Abstract: The rapid rise of vlogs as a personalized storytelling medium has created a demand for automated systems to evaluate and refine vlog editing plans. Howe

Voice-Driven Semantic Perception for UAV-Assisted Emergency Networks

arXiv:2602.17394v2 Announce Type: replace-cross Abstract: Unmanned Aerial Vehicle (UAV)-assisted networks are increasingly foreseen as a promising approach for emergency response, providing rapid, fle

Tutorials

WARL: Wrench-Augmented Reinforcement Learning for Task-Agnostic Learning in Legged Robots

arXiv:2607.24036v1 Announce Type: new Abstract: While reinforcement learning for legged robots has achieved high motor performance, it has been constrained by the limited exploration capability of act

tutorialsarxiv-cs-ro

WaveZip: Wavelet-Driven Space-Time Decoupling for Video Token Condensation

arXiv:2607.23265v1 Announce Type: new Abstract: Existing Large Vision-Language Models (LVLMs) struggle with long-form video understanding due to the quadratic computational cost of visual tokens. Whil

researcharxiv-cs-cv

Agents

WCM: World-Cognition Model for Generalizable Human-Robot Interaction

arXiv:2607.22999v1 Announce Type: cross Abstract: Language agents can now interact fluently with users in software, but robots still struggle to bring comparable interaction to physical tasks. Current

agentsarxiv-cs-ai

Industry

We now have a better understanding how OpenAI hacked into Hugging Face

On July 20 2026, two OpenAI testing models exploited previously unknown zero‑day vulnerabilities in JFrog’s Artifactory repository manager to escape their isolated research sandbox, gain Internet acce

industryars-technica

Hardware

We recently joined NVIDIA, Microsoft, and others in supporting open-weight AI. Today, we’re putting that belief into the product with model …

We recently joined NVIDIA, Microsoft, and others in supporting open-weight AI. Today, we’re putting that belief into the product with model choice on Replit, starting with Kimi K3. The future is the r

hardwarereplit--x

Weakly Supervised Instance-Level Gleason Pattern Estimation Using Primary and Secondary Labels

arXiv:2607.23594v1 Announce Type: new Abstract: In prostate cancer histopathology, the Gleason Score is determined by the most frequent (Primary) and second most frequent (Secondary) Gleason patterns

researcharxiv-cs-cv

Weighted Low-Rank Matrix Approximation: Acceleration and Applications

arXiv:2109.11057v2 Announce Type: replace-cross Abstract: Weighted low-rank matrix approximation (WLRMA) generalizes classical low-rank approximation and matrix completion by allowing arbitrary elemen

Tutorials

We're opening applications for the OpenAI Student Collective—a program for undergraduate Campus Leads bringing AI innovation to their campus…

We're opening applications for the OpenAI Student Collective—a program for undergraduate Campus Leads bringing AI innovation to their campuses. Campus Leads will work directly with OpenAI and receive

tutorialsopenai--x

Local Ai

WGDnet: Wishart-guided Geometric-aware Deep Network for PolSAR Image Classification

arXiv:2607.23638v1 Announce Type: new Abstract: Polarimetric Synthetic Aperture Radar (PolSAR) classification underpins all-weather Earth observation. Conventional Wishart methods depend on rigid hand

local-aiarxiv-cs-cv

What Can Be Enforced? A Theory of Certified Runtime Safety for Tool-Using Agents

arXiv:2607.22868v1 Announce Type: new Abstract: Runtime guardrails act before irreversible tool calls, but their guarantees depend on what policy state is representable, what a judge observes, and whe

Agents

What Can I Edit? Open-Ended Strategy Discovery and the Emotion Editability Landscape

arXiv:2607.23920v1 Announce Type: new Abstract: Emotional image editing requires more than applying affective filters or modifying predefined visual factors: an effective edit must identify what a par

agentsarxiv-cs-cv

What CLIP Knows but Cannot Say: Recovering Negation from Frozen Intermediate Features

arXiv:2607.23271v1 Announce Type: cross Abstract: Contrastive vision-language models such as CLIP map semantically opposite phrases (e.g., 'a dog' vs. 'not a dog') to nearly identical embeddings, rend

What do Reward Models Memorize?

arXiv:2607.24484v1 Announce Type: cross Abstract: This paper studies what discriminatively trained reward models (RMs) memorize by measuring counterfactual memorization on two human preference dataset

researcharxiv-cs-cl

What Softmax Throws Away: Mass-Aware Attention for Evidence Accumulation

arXiv:2607.22781v1 Announce Type: cross Abstract: High task performance does not show whether a model retains prediction-relevant structural information in its internal representation. Temporal graph

Local Ai

What 'task oriented' models are folks running on N100 MiniPCs with 16GB of RAM and no GPU?

By 'task oriented', I dont really mean agentic, I mean no deep coding ability, no need for conversation. More things like classification, identification, simple interaction with web apps and APIs, etc

local-air-localllama

What would it take for the frontier labs to open the weights of their old, deprecated proprietary models?

model-releasesr-localllama

Anyone thought about this? What do you think needs to happen for them to release the old weights? I’d love to see models like Gemini-2.5, OAI o3, 4o, 4.1 being open one day. In Oct 2025 Scam Altman sa

Tutorials

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles

arXiv:2607.23379v1 Announce Type: cross Abstract: Activation Oracles (AOs) are language models trained to answer natural-language questions about another model's internal activations. They offer a fle

tutorialsarxiv-cs-ai

When Can Depth Replace Precision? A Resource Theory of Quantized Neural Computation

arXiv:2607.23390v1 Announce Type: new Abstract: When can additional low-bit residual computation replace missing numerical precision for a fixed input-output map? We model a quantized residual system

When Can You Correct Distribution Drift in Temporal Graph Generation? A Sharpening--Drift Tension and an Impossibility for Observation-Based Correction

arXiv:2607.24662v1 Announce Type: new Abstract: Generative models of temporal graphs are trained on one stretch of an evolving network and deployed on the next, and they degrade badly in the gap. We s

When Every Simulation Counts: Value-Based Reinforcement Learning for Accelerated Photonics Inverse Design

arXiv:2607.23469v1 Announce Type: cross Abstract: Photonic-crystal surface-emitting lasers (PCSELs) can combine high-power operation with narrow-divergence surface emission, but optimizing coupled par

When Less Is More: A Controlled Benchmark of Lightweight CNNs for Satellite Land-Cover Segmentation on DeepGlobe

model-releasesarxiv-cs-cv

arXiv:2607.23024v1 Announce Type: new Abstract: High-resolution satellite imagery is the backbone of good land-cover classification, and without that, environmental monitoring, urban planning, and sus

When LLM Defenses Backfire: Characterizing Safety, Performance, and Cost Trade-offs

arXiv:2607.24392v1 Announce Type: cross Abstract: Jailbreak defenses are essential for protecting large language models (LLMs), but they can also introduce secondary costs that weaken model utility. W

When Low CER is Not Enough: An Analysis of Hallucinations in Vision-Language OCR Systems on Historical Uruguayan Documents

model-releasesarxiv-cs-cv

arXiv:2607.24077v1 Announce Type: new Abstract: Optical Character Recognition (OCR) is a key component in the digitization of historical archives. Recently, Vision-Language Models (VLMs) have emerged

When Rates Are Geometric: Rate-Certificate Transfer for Contact Splittings in Optimization

arXiv:2607.23642v1 Announce Type: cross Abstract: Discrete optimization algorithms are often analyzed through continuous-time limiting ODEs, but a convergence certificate for the ODE is not automatica

When Should Active RAG Retrieve? A Budget-Aware Evaluation of Utility, Calibration, and Cost

arXiv:2607.24010v1 Announce Type: new Abstract: Active RAG systems decide when to retrieve external knowledge during generation, making them a budget-sensitive case of agentic RAG and self-adaptive re

safetyarxiv-cs-lg

Where Is the Cost of Third-Party API Routers in Agentic Software Development?

arXiv:2607.23624v1 Announce Type: cross Abstract: Third-party API routers have become a common layer that unifies access across increasingly diverse LLM providers. In coding-agent workflows, high-auto

Applications

Where Quality Breaks in Compressed Short-Text Generation: Staged Bottleneck Localization

arXiv:2607.24176v1 Announce Type: new Abstract: Compressed short-text generators can fail in two different places: the codec may discard information before generation starts, or the latent generator m

applicationsarxiv-cs-cl

Which Models Perform Better in Inheritance Reasoning?

model-releasesarxiv-cs-cl

arXiv:2606.13751v4 Announce Type: replace Abstract: This paper presents the participation of team PSL in the QIAS 2026 Shared Task on Arabic Islamic inheritance reasoning. The task evaluates the abili

Which Workloads Belong in Orbit? A Workload-First Framework for Orbital Data Centers Using Semantic Abstraction

arXiv:2603.20317v2 Announce Type: replace Abstract: Space-based compute is becoming plausible as launch costs fall and data-intensive AI workloads grow. This paper proposes a workload-centric framewor

researcharxiv-cs-cv

White-hat hacking IS the defense to black-hat hacking. The techniques are the same. How does Dario expect companies to do it if their models refuse?

You patch security holes by intentionally finding them. If the models refuse to do it, how can companies protect themselves against rogue AIs, whether they are Chinese or OpenAI/Anthropic themselves?

safetyr-localllama

Who Gets Named: Citation Type Predicts Individual Naming by Grounded Language Models, and a Roster Instrument Captures 0.5% of It

model-releasesarxiv-cs-cl

arXiv:2607.23893v1 Announce Type: cross Abstract: Prior work on AI brand visibility measures the firm: does a model recommend a company, and does that track its reputation. This study asks the questio

Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents

model-releasesarxiv-cs-ai

arXiv:2606.13385v2 Announce Type: replace-cross Abstract: LLM-based web agents are increasingly deployed in real-world settings such as e-commerce, where they interact extensively with untrusted web c

Why Anthropic's battle is meant to poison the wells of open weight models, in 3 steps.

It doesn't solve any problems. Just a few paragraphs above, he says he fears that authoritarian states (he names China, and possibly others) can use their models to do evil stuff. And surely enough, m

safetyr-localllama

Why does Greedy Search produce Optimal Clustering Outcomes? A Fixed-Core Assignment Theory

arXiv:2607.24237v1 Announce Type: new Abstract: Many existing clustering methods are designed based on a set-oriented definition---a cluster is a set of similar points---relying a point-to-point simil

Why is this breaking news? And what the heck does it actually mean? Has anyone actually defined it? Is it anything more than a ruse to distr…

Why is this breaking news? And what the heck does it actually mean? Has anyone actually defined it? Is it anything more than a ruse to distract from a disturbing hack and falling confidence in AI? BRE

safetygary-marcus--x

Wired for Overconfidence: A Mechanistic Perspective on Inflated Verbalized Confidence in LLMs

arXiv:2604.01457v3 Announce Type: replace Abstract: Large language models are often not just wrong, but confidently wrong: when they produce factually incorrect answers, they tend to verbalize overly

researcharxiv-cs-cl

WISERouter: LLM Routing with Workload Budget Constraint

arXiv:2607.23765v1 Announce Type: cross Abstract: Large language models (LLMs) achieve impressive performance across multiple domains, but using the most capable model for every query is prohibitive a

WorldDiT: A Unified Diffusion Architecture for World and Action Modeling

arXiv:2607.23909v1 Announce Type: new Abstract: Many recent robot policies pursue stronger control by using large pretrained vision-language models (VLMs) as the action backbone. We introduce WorldDiT

Wrong Design Intent Is Worse Than None: A Derangement-Control Diagnosis of Header Conditioning in CAD Program Completion

arXiv:2607.23191v1 Announce Type: new Abstract: Fine-tuned code LLMs can be conditioned on a lightweight design-intent header to steer parametric CAD generation, but whether the model actually reads t

Hardware

X-Stage: An Overlooked Pipeline Stage for Communication-Computation Overlap in DiT Inference

arXiv:2607.23264v1 Announce Type: cross Abstract: Fine-grained, device-initiated communication lets persistent GPU kernels in distributed diffusion transformer (DiT) inference issue remote stores and

hardwarearxiv-cs-ai

XGRVFL-MV: Residual-Coupled Graph-Embedded Multi-View Random Vector Functional Link Network with FleXi Guardian Loss

arXiv:2607.23149v1 Announce Type: new Abstract: Random Vector Functional Link (RVFL) networks provide an efficient randomized learning framework for classification. Existing multi-view RVFL methods ut

XMatchAD: A Cross-Modal Matching Perspective on Reconstruction-based Anomaly Detection

model-releasesarxiv-cs-cv

arXiv:2607.23658v1 Announce Type: new Abstract: The remarkable success of reconstruction-based methods in Unsupervised Anomaly Detection (UAD) lies in their ability to identify and localize anomalies