research

When Stronger Triggers Backfire: A High-Dimensional Theory of Backdoor Attacks

arXiv:2605.22481v1 Announce Type: new Abstract: Backdoor poisoning attacks behave counter-intuitively in high dimensions: stronger training triggers can help the defender. We study regularised general

researcharxiv-cs-lg

23 May 2026

Why?

Why? Most green-card applicants will need to go abroad to apply for permanent residency at an American consulate, rather than filing from within the U.S. as they do now, the Trump administration annou

researchyann-lecun--x

23 May 2026

Winner-Take-All bottlenecks enforce disentangled symbolic representations in multi-task learning

arXiv:2605.22472v1 Announce Type: new Abstract: Winner-take-all (WTA) networks constitute a central circuit motif in cortical networks of the brain. In addition, WTA-like activations are abundant in m

researcharxiv-cs-lg

23 May 2026

Zero-shot adaptation to order book dynamics

arXiv:2605.21707v1 Announce Type: cross Abstract: We describe an adaptive market-making architecture that preserves the analytical structure of the Avellaneda--Stoikov framework while introducing a su

researcharxiv-cs-lg

23 May 2026

4D-GSW: Kinematic-Aware Spatio-Temporal Consistent Watermarking for 4D Gaussian Splatting

arXiv:2605.22342v1 Announce Type: new Abstract: While 4D Gaussian Splatting (4DGS) has revolutionized high-fidelity dynamic reconstruction, safeguarding the intellectual property of these assets remai

A Robust Semantic Segmentation Pipeline for the CVPR 2026 8th UG2+ Challenge Track 2

arXiv:2605.22216v1 Announce Type: new Abstract: This report presents our solution for the WeatherProof Dataset Challenge, namely CVPR 2026 8th UG2+ Challenge Track 2: Semantic Segmentation in Adverse

Ablate-to-Validate: Are Vision-Language Models Really Using Continuous Thought Tokens?

arXiv:2605.21642v1 Announce Type: new Abstract: Vision-language models (VLMs) are increasingly augmented with continuous or latent non-textual tokens intended to support 'visual thinking.' Despite imp

Accelerated Test-Time Scaling with Model-Free Speculative Sampling

arXiv:2506.04708v3 Announce Type: replace Abstract: Language models have demonstrated remarkable capabilities in reasoning tasks through test-time scaling techniques like best-of-N sampling and tree s

Accelerating Vision Foundation Models with Drop-in Depthwise Convolution

arXiv:2605.22132v1 Announce Type: new Abstract: Pretrained vision foundation models deliver strong performance across tasks with limited fine-tuning. However, their Vision Transformer (ViT) backbones

Access Paths for Efficient Ordering with Large Language Models

arXiv:2509.00303v3 Announce Type: replace-cross Abstract: In this work, we present the exttt{LLM ORDER BY} semantic operator as a logical abstraction and conduct a systematic study of its physical imp

researcharxiv-cs-ai

Action with Visual Primitives

arXiv:2605.22183v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have emerged as a promising paradigm for generalist robotic manipulation. A common design in current architectures m

researcharxiv-cs-ro

An Evidence Hierarchy for Bayesian Object Classification via OSINT-Aided Heterogeneous Sensor Fusion

arXiv:2605.22259v1 Announce Type: cross Abstract: Heterogeneous sensor fusion is vital for detecting, localizing, and classifying CBRNE threats. However, individual sensors are often only capable of d

An Open Multi-Center Whole-Body FDG PET/CT Foundation Model for Tumor Segmentation

arXiv:2605.21835v1 Announce Type: cross Abstract: The synergistic interpretation of anatomical information from computed tomography (CT) and metabolic information from positron emission tomography (PE

Analytical and Experimental Force Analysis of a Soft Linear Pneumatic Actuator

arXiv:2605.21836v1 Announce Type: new Abstract: Soft sleeve actuators (SSAs) have recently been developed as a pneumatic actuation approach for wearable and assistive robotic systems. By integrating t

researcharxiv-cs-ro

Assisted Counterspeech Writing at the Crossroads of Hate Speech and Misinformation

arXiv:2605.22435v1 Announce Type: new Abstract: Hate speech and misinformation frequently co-occur online, amplifying prejudice and polarization. Given their scale, using Large Language Models (LLMs)

Attacking the Spike: On the Transferability and Security of Spiking Neural Networks to Adversarial Examples

arXiv:2209.03358v5 Announce Type: replace-cross Abstract: Spiking neural networks (SNNs) have attracted much attention for their high energy efficiency and recent advances in classification performanc

Bernini: Latent Semantic Planning for Video Diffusion

arXiv:2605.22344v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) and diffusion models have each reached remarkable maturity: MLLMs excel at reasoning over heterogeneous multimo

Broken Memories: Detecting and Mitigating Memorization in Diffusion Models with Degraded Generations

arXiv:2605.22050v1 Announce Type: new Abstract: While diffusion models excel at generating high-quality images, their tendency to memorize training data poses significant privacy and copyright risks.

Cambrian-P: Pose-Grounded Video Understanding

arXiv:2605.22819v1 Announce Type: new Abstract: Camera pose matters. The position and orientation of each viewpoint define a shared spatial coordinate frame that relates observations across video fram

Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation

arXiv:2602.02214v3 Announce Type: replace Abstract: To achieve real-time interactive video generation, current methods distill pretrained bidirectional video diffusion models into few-step autoregress

Cell Phantom Video Generation in Elliptical Fourier Descriptor Domain

arXiv:2605.22563v1 Announce Type: new Abstract: Training Deep Neural Networks for tracking individual cells in biomedical videos requires a large amount of annotated data. The annotation of videos for

Chinese sensorimotor and embodiment norms for 3,000 lexicalized concepts

arXiv:2605.22616v1 Announce Type: new Abstract: Understanding how conceptual knowledge is grounded in bodily experience, and to what extent machine systems can acquire such knowledge without direct se

Claim-Selective Certification for High-Risk Medical Retrieval-Augmented Generation

arXiv:2605.21949v1 Announce Type: new Abstract: Medical RAG systems in high-risk QA settings are often evaluated through a single answer-or-abstain decision, but mixed evidence may support one claim,

ConvNeXt-FD: A Fractal-Based Deep Model for Robust Biomedical Image Segmentation

arXiv:2605.22002v1 Announce Type: new Abstract: Biomedical image segmentation is a critical task in medical diagnosis and treatment planning, enabling precise delineation of anatomical structures and

CritiSense: Critical Digital Literacy and Resilience Against Misinformation

arXiv:2603.16672v2 Announce Type: replace-cross Abstract: Misinformation on social media undermines informed decision-making and public trust. Prebunking offers a proactive complement by helping users

D3Seg: Dependency-Aware Diffusion for Brain Tumor Segmentation with Missing Modalities

arXiv:2605.22249v1 Announce Type: new Abstract: Accurate brain tumor segmentation using multiparametric MRI is critical for effective treatment planning. However, in clinical settings, complete acquis

Decoupling Ego-Motion from Target Dynamics via Dual-Interval Motion Cues for UAV Detection

arXiv:2605.22605v1 Announce Type: cross Abstract: Object detection from Unmanned Aerial Vehicles (UAVs) is challenged by severe ego-motion, camera jitter, and large scale variations. While modern dete

DecQ: Detail-Condensing Queries for Enhanced Reconstruction and Generation in Representation Autoencoders

arXiv:2605.22777v1 Announce Type: new Abstract: Representation Autoencoders (RAEs) leverage frozen vision foundation models (VFMs) as tokenizer encoders, providing robust high-level representations th

DeferMem: Query-Time Evidence Distillation via Reinforcement Learning for Long-Term Memory QA

arXiv:2605.22411v1 Announce Type: new Abstract: Large language model (LLM) agents still struggle with long-term memory question answering, where answer-supporting evidence is often scattered across lo

Designing Conversations with the Dead: How People Engage with Generative Ghosts

arXiv:2605.21390v1 Announce Type: cross Abstract: We examine how people experience two choices in the design of generative ghosts, AI systems that are trained on data of the dead: representation, wher

researcharxiv-cs-ai

Detecting Synthetic Political Narratives in Cross-Platform Social Media Discourse

arXiv:2605.21540v1 Announce Type: cross Abstract: The proliferation of large language models has introduced a new paradigm of synthetic political communication in which narratives may be generated, se

Detecting Trojaned DNNs via Spectral Regression Analysis

arXiv:2605.21146v1 Announce Type: cross Abstract: Modern DNNs are repeatedly fine-tuned to incorporate new data and functionality. This evolutionary workflow introduces a security risk when updated da

researcharxiv-cs-ai

Detection of Virus and Small Cell Patches in Foci Images Using Switchable Convolution and Feature Pyramid Networks

arXiv:2605.22290v1 Announce Type: new Abstract: Accurate detection and counting of virus patches in focus-forming unit (FFU) images, also known as foci images, are important for quantifying viral infe

Direct content-based retrieval from music scores images

arXiv:2605.22255v1 Announce Type: new Abstract: The digitization of musical scores plays a crucial role in their preservation and accessibility, yet information retrieval still depends mainly on metad

Do Factual Recall Mechanisms Carry over from Text to Speech in Multimodal Language Models?

arXiv:2605.22170v1 Announce Type: new Abstract: In recent years, several Speech Language Models (SLMs) that represent speech and written text jointly have been presented. The question then emerges abo

Enhancing Causal Reasoning in Large Language Models: A Causal Attribution Model for Precision Fine-Tuning

arXiv:2401.00139v3 Announce Type: replace-cross Abstract: This paper introduces a causal attribution model to enhance the interpretability of large language models (LLMs) and improve their causal reas

Enhancing Gaze Reasoning in Vision Foundation Models for Gaze Following

arXiv:2605.22607v1 Announce Type: new Abstract: Gaze following requires both scene understanding and gaze reasoning to localize the gaze target of an in-scene person. Recently, vision foundation model

Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding

arXiv:2605.22078v1 Announce Type: cross Abstract: Recent advances in Multimodal Large Language Models (MLLMs) have significantly advanced video understanding tasks, yet challenges remain in efficientl

EntmaxKV: Support-Aware Decoding for Entmax Attention

arXiv:2605.21649v1 Announce Type: cross Abstract: Long-context decoding is increasingly limited by KV-cache memory traffic since each generated token attends over a cache whose size grows linearly wit

Entropy-Guided Self-Supervised Learning for Medical Image Classification

arXiv:2605.21970v1 Announce Type: cross Abstract: Accurate and robust medical image classification is paramount for early disease diagnosis and treatment planning. However, challenges such as limited

Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings

arXiv:2605.22391v1 Announce Type: cross Abstract: We present Epicure, a family of three sibling skip-gram ingredient embeddings retrained from scratch on a multilingual recipe corpus. We aggregate 4.1

Evaluation of Chunking Strategies for Effective Text Embedding in Low-Resource Language on Agricultural Documents

arXiv:2605.22203v1 Announce Type: new Abstract: In this study, we compare the performance of four text chunking approaches: Recursive, Khmer-Aware, Sentence-Based, and LLM-Based within a Retrieval-Aug

EvoScene-VLA: Evolving Scene Beliefs Inside the Action Decoder for Chunked Robot Control

arXiv:2605.21862v1 Announce Type: new Abstract: Chunked vision-language-action (VLA) policies predict multi-step robot controls, conditioning each update on the current visual observation alone. Yet r

researcharxiv-cs-ro

Faithful-MR1: Faithful Multimodal Reasoning via Anchoring and Reinforcing Visual Attention

arXiv:2605.22072v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has emerged as a promising paradigm for advancing complex reasoning in large language models, and

Flow-based Gaussian Splatting for Continuous-Scale Remote Sensing Image Super-Resolution

arXiv:2605.22147v1 Announce Type: new Abstract: High-resolution remote sensing images (RSIs) are crucial for Earth observation applications, yet acquiring them is often limited by sensor constraints a

Foresee-to-Ground: From Predictive Temporal Perception to Evidence-Driven Reasoning for Video Temporal Grounding

arXiv:2605.21973v1 Announce Type: new Abstract: Current Video-LLM approaches for Video Temporal Grounding (VTG) typically rely on direct timestamp generation from an unstructured visual-token stream,

ForeSplat: Optimization-Aware Foresight for Feed-Forward 3D Gaussian Splatting

arXiv:2605.22020v1 Announce Type: new Abstract: Feed-forward 3D Gaussian Splatting (3DGS) models offer fast single-pass reconstruction,but scaling them to match per-scene optimization quality is funda

From Baseline to Follow-Up: Counterfactual Spine DXA Image Synthesis in UK Biobank Using a Causal Hierarchical Variational Autoencoder

arXiv:2605.22649v1 Announce Type: new Abstract: Dual-energy X-ray absorptiometry (DXA) is widely used for large-scale skeletal assessment, yet learning controllable and interpretable factor-specific a