Search: “papers”

ZoomR: Memory Efficient Reasoning through Multi-Granularity Key Value Retrieval

arXiv:2604.10898v1 Announce Type: new Abstract: Large language models (LLMs) have shown great performance on complex reasoning tasks but often require generating long intermediate thoughts before reac

14 Apr 2026

Research

2D or 3D: Who Governs Salience in VLA Models? -- Tri-Stage Token Pruning Framework with Modality Salience Awareness

DGX agent

arXiv:2604.09244v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have emerged as the mainstream of embodied intelligence. Recent VLA models have expanded their input modalities fr

researcharxiv-cs-cv

13 Apr 2026

Adaptive Planning for Multi-Attribute Controllable Summarization with Monte Carlo Tree Search

arXiv:2509.26435v2 Announce Type: replace-cross Abstract: Controllable summarization moves beyond generic outputs toward human-aligned summaries guided by specified attributes. In practice, the interd

Adaptive Rigor in AI System Evaluation using Temperature-Controlled Verdict Aggregation via Generalized Power Mean

arXiv:2604.08595v1 Announce Type: cross Abstract: Existing evaluation methods for LLM-based AI systems, such as LLM-as-a-Judge, verdict systems, and NLI, do not always align well with human assessment

Adaptive Tuning of Parameterized Traffic Controllers via Multi-Agent Reinforcement Learning

arXiv:2512.07417v2 Announce Type: replace Abstract: Effective traffic control is essential for mitigating congestion in transportation networks. Conventional traffic management strategies, including r

AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society

arXiv:2502.08691v2 Announce Type: replace-cross Abstract: Understanding human behavior and society is a central focus in social sciences, with the rise of generative social science marking a significa

Anchored Sliding Window: Toward Robust and Imperceptible Linguistic Steganography

model-releasesarxiv-cs-cl

arXiv:2604.09066v1 Announce Type: new Abstract: Linguistic steganography based on language models typically assumes that steganographic texts are transmitted without alteration, making them fragile to

Attention-Based Sampler for Diffusion Language Models

arXiv:2604.08564v1 Announce Type: new Abstract: Auto-regressive models (ARMs) have established a dominant paradigm in language modeling. However, their strictly sequential decoding paradigm imposes fu

researcharxiv-cs-cl

B-MoE: A Body-Part-Aware Mixture-of-Experts 'All Parts Matter' Approach to Micro-Action Recognition

arXiv:2603.24245v3 Announce Type: replace Abstract: Micro-actions, fleeting and low-amplitude motions, such as glances, nods, or minor posture shifts, carry rich social meaning but remain difficult fo

Balancing User Preferences by Social Networks: A Condition-Guided Social Recommendation Model for Mitigating Popularity Bias

arXiv:2405.16772v2 Announce Type: replace-cross Abstract: Social recommendation models weave social interactions into their design to provide uniquely personalized recommendation results for users. Ho

safetyarxiv-cs-lg

Beyond Flicker: Detecting Kinematic Inconsistencies for Generalizable Deepfake Video Detection

arXiv:2512.04175v2 Announce Type: replace Abstract: Generalizing deepfake detection to unseen manipulations remains a key challenge. A recent approach to tackle this issue is to train a network with p

Boosting Brain-inspired Path Integration Efficiency via Learning-based Replication of Continuous Attractor Neurodynamics

arXiv:2511.17687v2 Announce Type: replace Abstract: The brain's Path Integration (PI) mechanism offers substantial guidance and inspiration for Brain-Inspired Navigation (BIN). However, the PI capabil

Bridging SFT and RL: Dynamic Policy Optimization for Robust Reasoning

arXiv:2604.08926v1 Announce Type: new Abstract: Post-training paradigms for Large Language Models (LLMs), primarily Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), face a fundamental dil

safetyarxiv-cs-lg

Building Better Environments for Autonomous Cyber Defence

arXiv:2604.08805v1 Announce Type: cross Abstract: In November 2025, the authors ran a workshop on the topic of what makes a good reinforcement learning (RL) environment for autonomous cyber defence (A

agentsarxiv-cs-ai

CaRLi-V: Camera-RADAR-LiDAR Point-Wise 3D Velocity Estimation

arXiv:2511.01383v2 Announce Type: replace Abstract: Accurate point-wise velocity estimation in 3D is crucial for robot interaction with non-rigid dynamic agents, enabling robust performance in path pl

researcharxiv-cs-ro

Characterizing Lidar Range-Measurement Ambiguity due to Multiple Returns

arXiv:2604.09282v1 Announce Type: cross Abstract: Reliable position and attitude sensing is critical for highly automated vehicles that operate on conventional roadways. Lidar sensors are increasingly

Constraining Sequential Model Editing with Editing Anchor Compression

arXiv:2503.00035v2 Announce Type: replace-cross Abstract: Large language models (LLMs) struggle with hallucinations due to false or outdated knowledge. Given the high resource demands of retraining th

Deep Light Pollution Removal in Night Cityscape Photographs

arXiv:2604.09145v1 Announce Type: new Abstract: Nighttime photography is severely degraded by light pollution induced by pervasive artificial lighting in urban environments. After long-range scatterin

Tutorials

Degradation-Robust Fusion: An Efficient Degradation-Aware Diffusion Framework for Multimodal Image Fusion in Arbitrary Degradation Scenarios

arXiv:2604.08922v1 Announce Type: new Abstract: Complex degradations like noise, blur, and low resolution are typical challenges in real world image fusion tasks, limiting the performance and practica

tutorialsarxiv-cs-cv

Dejavu: Towards Experience Feedback Learning for Embodied Intelligence

arXiv:2510.10181v3 Announce Type: replace-cross Abstract: Embodied agents face a fundamental limitation: once deployed in real-world environments, they cannot easily acquire new knowledge to improve t

Demystifying the Silence of Correctness Bugs in PyTorch Compiler

arXiv:2604.08720v1 Announce Type: cross Abstract: Performance optimization of AI infrastructure is key to the fast adoption of large language models (LLMs). The PyTorch compiler (torch.compile), a cor

researcharxiv-cs-ai

Detecting Diffusion-generated Images via Dynamic Assembly ForestsDetecting Diffusion-generated Images via Dynamic Assembly Forests

arXiv:2604.09106v1 Announce Type: new Abstract: Diffusion models are known for generating high-quality images, causing serious security concerns. To combat this, most efforts rely on deep neural netwo

Detection of Hate and Threat in Digital Forensics: A Case-Driven Multimodal Approach

arXiv:2604.08609v1 Announce Type: cross Abstract: Digital forensic investigations increasingly rely on heterogeneous evidence such as images, scanned documents, and contextual reports. These artifacts

researcharxiv-cs-ai

Do We Really Need to Approach the Entire Pareto Front in Many-Objective Bayesian Optimisation?

arXiv:2604.09417v1 Announce Type: new Abstract: Many-objective optimisation, a subset of multi-objective optimisation, involves optimisation problems with more than three objectives. As the number of

Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight

model-releasesarxiv-cs-ro

arXiv:2501.14377v2 Announce Type: replace Abstract: Autonomous drone racing has risen as a challenging robotic benchmark for testing the limits of learning, perception, planning, and control. Expert h

Enhancing the Safety of Medical Vision-Language Models by Synthetic Demonstrations

arXiv:2506.09067v2 Announce Type: replace-cross Abstract: Generative medical vision-language models~(Med-VLMs) are primarily designed to generate complex textual information~(e.g., diagnostic reports)

Exploring Cross-lingual Latent Transplantation: Mutual Opportunities and Open Challenges

arXiv:2412.12686v3 Announce Type: replace Abstract: Current large language models (LLMs) often exhibit imbalances in multilingual capabilities and cultural adaptability, largely attributed to their En

researcharxiv-cs-cl

Exploring Teachers' Perspectives on Using Conversational AI Agents for Group Collaboration

arXiv:2602.07142v2 Announce Type: replace-cross Abstract: Collaboration is a cornerstone of 21st-century learning, yet teachers continue to face challenges in supporting productive peer interaction. E

FashionStylist: An Expert Knowledge-enhanced Multimodal Dataset for Fashion Understanding

model-releasesarxiv-cs-cv

arXiv:2604.09249v1 Announce Type: new Abstract: Fashion understanding requires both visual perception and expert-level reasoning about style, occasion, compatibility, and outfit rationale. However, ex

Finite-Sample Analysis of Nonlinear Independent Component Analysis:Sample Complexity and Identifiability Bounds

arXiv:2604.08850v1 Announce Type: new Abstract: Independent Component Analysis (ICA) is a fundamental unsupervised learning technique foruncovering latent structure in data by separating mixed signals

FIRE-CIR: Fine-grained Reasoning for Composed Fashion Image Retrieval

model-releasesarxiv-cs-cv

arXiv:2604.09114v1 Announce Type: new Abstract: Composed image retrieval (CIR) aims to retrieve a target image that depicts a reference image modified by a textual description. While recent vision-lan

Fisher-Geometric Diffusion in Stochastic Gradient Descent: Optimal Rates, Oracle Complexity, and Information-Theoretic Limits

arXiv:2603.02417v3 Announce Type: replace-cross Abstract: Classical stochastic-approximation analyses treat the covariance of stochastic gradients as an exogenous modeling input. We show that under ex

researcharxiv-cs-lg

FIT-GNN: Faster Inference Time for GNNs that 'FIT' in Memory Using Coarsening

arXiv:2410.15001v5 Announce Type: replace Abstract: Scalability of Graph Neural Networks (GNNs) remains a significant challenge. To tackle this, methods like coarsening, condensation, and computation

From Selection to Scheduling: Federated Geometry-Aware Correction Makes Exemplar Replay Work Better under Continual Dynamic Heterogeneity

arXiv:2604.08617v1 Announce Type: cross Abstract: Exemplar replay has become an effective strategy for mitigating catastrophic forgetting in federated continual learning (FCL) by retaining representat

GAN-Enhanced Deep Reinforcement Learning for Semantic-Aware Resource Allocation in 6G Network Slicing

arXiv:2604.08576v1 Announce Type: cross Abstract: Sixth-generation (6G) wireless networks must support heterogeneous services: enhanced Mobile Broadband (eMBB) requiring 1 Tbps data rates, massive Mac

Generative Simulation for Policy Learning in Physical Human-Robot Interaction

arXiv:2604.08664v1 Announce Type: new Abstract: Developing autonomous physical human-robot interaction (pHRI) systems is limited by the scarcity of large-scale training data to learn robust robot beha

safetyarxiv-cs-ro

GeRM: A Generative Rendering Model From Physically Realistic to Photorealistic

arXiv:2604.09304v1 Announce Type: new Abstract: For decades, Physically-Based Rendering (PBR) is the fundation of synthesizing photorealisitic images, and therefore sometimes roughly referred as Photo

agentsarxiv-cs-cv

GNN-as-Judge: Unleashing the Power of LLMs for Graph Learning with GNN Feedback

arXiv:2604.08553v1 Announce Type: cross Abstract: Large Language Models (LLMs) have shown strong performance on text-attributed graphs (TAGs) due to their superior semantic understanding ability on te

GRM: Utility-Aware Jailbreak Attacks on Audio LLMs via Gradient-Ratio Masking

arXiv:2604.09222v1 Announce Type: cross Abstract: Audio large language models (ALLMs) enable rich speech-text interaction, but they also introduce jailbreak vulnerabilities in the audio modality. Exis

How Should Video LLMs Output Time? An Analysis of Efficient Temporal Grounding Paradigms

model-releasesarxiv-cs-cv

arXiv:2604.08966v1 Announce Type: new Abstract: While Multimodal Large Language Models (MLLMs) have advanced Video Temporal Grounding (VTG), existing methods often couple output paradigms with differe

Hypergraph Neural Networks Accelerate MUS Enumeration

arXiv:2604.09001v1 Announce Type: new Abstract: Enumerating Minimal Unsatisfiable Subsets (MUSes) is a fundamental task in constraint satisfaction problems (CSPs). Its major challenge is the exponenti

agentsarxiv-cs-ai

Imitation Learning for Combinatorial Optimisation under Uncertainty

arXiv:2601.05383v4 Announce Type: replace Abstract: Imitation learning (IL) provides a data-driven framework for approximating policies for large-scale combinatorial optimisation problems formulated a

researcharxiv-cs-lg

Implicit Bias in Deep Linear Discriminant Analysis

arXiv:2603.02622v2 Announce Type: replace Abstract: While the Implicit Bias(or Implicit Regularization) of standard loss functions has been studied, the optimization geometry induced by discriminative

safetyarxiv-cs-lg

InsEdit: Towards Instruction-based Visual Editing via Data-Efficient Video Diffusion Models Adaptation

arXiv:2604.08646v1 Announce Type: new Abstract: Instruction-based video editing is a natural way to control video content with text, but adapting a video generation model into an editor usually appear

Interactive ASR: Towards Human-Like Interaction and Semantic Coherence Evaluation for Agentic Speech Recognition

arXiv:2604.09121v1 Announce Type: cross Abstract: Recent years have witnessed remarkable progress in automatic speech recognition (ASR), driven by advances in model architectures and large-scale train

agentsarxiv-cs-ai

Intrinsic Concept Extraction Based on Compositional Interpretability

arXiv:2603.11795v2 Announce Type: replace Abstract: Unsupervised Concept Extraction aims to extract concepts from a single image; however, existing methods suffer from the inability to extract composa

Applications

Investigating Multimodal Large Language Models to Support Usability Evaluation

arXiv:2508.16165v2 Announce Type: replace-cross Abstract: Usability evaluation is an essential method to support the design of effective and intuitive user interfaces (UIs). However, it commonly relie

applicationsarxiv-cs-ai

Tutorials

Learning General Representation of 12-Lead Electrocardiogram with a Joint-Embedding Predictive Architecture

arXiv:2410.08559v5 Announce Type: replace-cross Abstract: Electrocardiogram (ECG) captures the heart's electrical signals, offering valuable information for diagnosing cardiac conditions. However, the

tutorialsarxiv-cs-ai