Search: “framework”

arXiv:2604.21231v1 Announce Type: cross Abstract: Efficient inference for on-device Large Language Models (LLMs) remains challenging due to limited hardware resources and the high cost of the prefill

local-aiarxiv-cs-ai

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

Spectral Embeddings Leak Graph Topology: Theory, Benchmark, and Adaptive Reconstruction

DGX agent

arXiv:2604.21094v1 Announce Type: new Abstract: Graph Neural Networks (GNNs) excel on relational data, but standard benchmarks unrealistically assume the graph is centrally available. In practice, set

model-releasesarxiv-cs-lg

24 Apr 2026

Stealthy Backdoor Attacks against LLMs Based on Natural Style Triggers

arXiv:2604.21700v1 Announce Type: cross Abstract: The growing application of large language models (LLMs) in safety-critical domains has raised urgent concerns about their security. Many recent studie

Strategic Heterogeneous Multi-Agent Architecture for Cost-Effective Code Vulnerability Detection

model-releasesarxiv-cs-lg

arXiv:2604.21282v1 Announce Type: cross Abstract: Automated code vulnerability detection is critical for software security, yet existing approaches face a fundamental trade-off between detection accur

StructMem: Structured Memory for Long-Horizon Behavior in LLMs

arXiv:2604.21748v1 Announce Type: cross Abstract: Long-term conversational agents need memory systems that capture relationships between events, not merely isolated facts, to support temporal reasonin

researcharxiv-cs-ai

SurgViVQA: Temporally-Grounded Video Question Answering for Surgical Scene Understanding

model-releasesarxiv-cs-cv

arXiv:2511.03325v3 Announce Type: replace Abstract: Video Question Answering (VideoQA) in the surgical domain aims to enhance intraoperative understanding by enabling AI models to reason over temporal

Survey on Evaluation of LLM-based Agents

arXiv:2503.16416v2 Announce Type: replace Abstract: LLM-based agents represent a paradigm shift in AI, enabling autonomous systems to plan, reason, and use tools while interacting with dynamic environ

Synthetic Data in Education: Empirical Insights from Traditional Resampling and Deep Generative Models

arXiv:2604.21031v1 Announce Type: cross Abstract: Synthetic data generation offers promise for addressing data scarcity and privacy concerns in educational technology, yet practitioners lack empirical

TEMA: Anchor the Image, Follow the Text for Multi-Modification Composed Image Retrieval

model-releasesarxiv-cs-cv

arXiv:2604.21806v1 Announce Type: new Abstract: Composed Image Retrieval (CIR) is an important image retrieval paradigm that enables users to retrieve a target image using a multimodal query that cons

Temporal Prototyping and Hierarchical Alignment for Unsupervised Video-based Visible-Infrared Person Re-Identification

arXiv:2604.21324v1 Announce Type: new Abstract: Visible-infrared person re-identification (VI-ReID) enables cross-modality identity matching for all-day surveillance, yet existing methods predominantl

safetyarxiv-cs-cv

Temporal Taskification in Streaming Continual Learning: A Source of Evaluation Instability

model-releasesarxiv-cs-lg

arXiv:2604.21930v1 Announce Type: new Abstract: Streaming Continual Learning (CL) typically converts a continuous stream into a sequence of discrete tasks through temporal partitioning. We argue that

The Effect of Idea Elaboration on the Automatic Assessment of Idea Originality

arXiv:2604.20569v1 Announce Type: cross Abstract: Automatic systems are increasingly used to assess the originality of responses in creative tasks. They offer a potential solution to key limitations o

Thinking Like a Botanist: Challenging Multimodal Language Models with Intent-Driven Chain-of-Inquiry

arXiv:2604.20983v1 Announce Type: cross Abstract: Vision evaluations are typically done through multi-step processes. In most contemporary fields, experts analyze images using structured, evidence-bas

Agents

Tool Attention Is All You Need

Tool Attention Is All You Need // Tool Attention Is All You Need // New research proposes a practical fix for the hidden 'MCP tax.' The work introduces a dynamic tool gating mechanism built on an Inte

agentsdair-ai--x

Towards a Systematic Risk Assessment of Deep Neural Network Limitations in Autonomous Driving Perception

arXiv:2604.20895v1 Announce Type: cross Abstract: Safety and security are essential for the admission and acceptance of automated and autonomous vehicles. Deep neural networks (DNNs) are widely used f

safetyarxiv-cs-lg

Tutorials

Transferable Physics-Informed Representations via Closed-Form Head Adaptation

arXiv:2604.21761v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) have garnered significant interest for their potential in solving partial differential equations (PDEs) that go

tutorialsarxiv-cs-lg

Transformer-Progressive Mamba Network for Lightweight Image Super-Resolution

arXiv:2511.03232v2 Announce Type: replace Abstract: Recently, Mamba-based super-resolution (SR) methods have demonstrated the ability to capture global receptive fields with linear complexity, address

researcharxiv-cs-cv

Transient Turn Injection: Exposing Stateless Multi-Turn Vulnerabilities in Large Language Models

arXiv:2604.21860v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly integrated into sensitive workflows, raising the stakes for adversarial robustness and safety. This pape

UAU-Net: Uncertainty-aware Representation Learning and Evidential Classification for Facial Action Unit Detection

arXiv:2604.21227v1 Announce Type: new Abstract: Facial action unit (AU) detection remains challenging because it involves heterogeneous, AU-specific uncertainties arising at both the representation an

researcharxiv-cs-cv

Understanding and Mitigating Spurious Signal Amplification in Test-Time Reinforcement Learning for Math Reasoning

arXiv:2604.21327v1 Announce Type: cross Abstract: Test-time reinforcement learning (TTRL) always adapts models at inference time via pseudo-labeling, leaving it vulnerable to spurious optimization sig

Unlocking the Power of Large Language Models for Multi-table Entity Matching

arXiv:2604.21238v1 Announce Type: new Abstract: Multi-table entity matching (MEM) addresses the limitations of dual-table approaches by enabling simultaneous identification of equivalent entities acro

researcharxiv-cs-cl

Using Machine Mental Imagery for Representing Common Ground in Situated Dialogue

arXiv:2604.21144v1 Announce Type: cross Abstract: Situated dialogue requires speakers to maintain a reliable representation of shared context rather than reasoning only over isolated utterances. Curre

Value-Conflict Diagnostics Reveal Widespread Alignment Faking in Language Models

arXiv:2604.20995v1 Announce Type: new Abstract: Alignment faking, where a model behaves aligned with developer policy when monitored but reverts to its own preferences when unobserved, is a concerning

VARestorer: One-Step VAR Distillation for Real-World Image Super-Resolution

arXiv:2604.21450v1 Announce Type: cross Abstract: Recent advancements in visual autoregressive models (VAR) have demonstrated their effectiveness in image generation, highlighting their potential for

VistaBot: View-Robust Robot Manipulation via Spatiotemporal-Aware View Synthesis

model-releasesarxiv-cs-ro

arXiv:2604.21914v1 Announce Type: new Abstract: Recently, end-to-end robotic manipulation models have gained significant attention for their generalizability and scalability. However, they often suffe

VLA-Forget: Vision-Language-Action Unlearning for Embodied Foundation Models

arXiv:2604.03956v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) models are emerging as embodied foundation models for robotic manipulation, but their deployment introduces a new

When Bigger Isn't Better: A Comprehensive Fairness Evaluation of Political Bias in Multi-News Summarisation

arXiv:2604.21309v1 Announce Type: new Abstract: Multi-document news summarisation systems are increasingly adopted for their convenience in processing vast daily news content, making fairness across d

safetyarxiv-cs-cl

When Prompts Override Vision: Prompt-Induced Hallucinations in LVLMs

arXiv:2604.21911v1 Announce Type: cross Abstract: Despite impressive progress in capabilities of large vision-language models (LVLMs), these systems remain vulnerable to hallucinations, i.e., outputs

Who Defines Fairness? Target-Based Prompting for Demographic Representation in Generative Models

arXiv:2604.21036v1 Announce Type: new Abstract: Text-to-image(T2I) models like Stable Diffusion and DALL-E have made generative AI widely accessible, yet recent studies reveal that these systems often

Applications

X2-N: A Transformable Wheel-legged Humanoid Robot with Dual-mode Locomotion and Manipulation

arXiv:2604.21541v1 Announce Type: new Abstract: Wheel-legged robots combine the efficiency of wheeled locomotion with the versatility of legged systems, enabling rapid traversal over both continuous a

applicationsarxiv-cs-ro

You Only Gaussian Once: Controllable 3D Gaussian Splatting for Ultra-Densely Sampled Scenes

model-releasesarxiv-cs-cv

arXiv:2604.21400v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) has revolutionized neural rendering, yet existing methods remain predominantly research prototypes ill-suited for productio

A Computational Model of Message Sensation Value in Short Video Multimodal Features that Predicts Sensory and Behavioral Engagement

arXiv:2604.19995v1 Announce Type: new Abstract: The contemporary media landscape is characterized by sensational short videos. While prior research examines the effects of individual multimodal featur

researcharxiv-cs-cv

A Vision-Language-Action Model for Adaptive Ultrasound-Guided Needle Insertion and Needle Tracking

arXiv:2604.20347v1 Announce Type: cross Abstract: Ultrasound (US)-guided needle insertion is a critical yet challenging procedure due to dynamic imaging conditions and difficulties in needle visualiza

ACT: Anti-Crosstalk Learning for Cross-Sectional Stock Ranking via Temporal Disentanglement and Structural Purification

arXiv:2604.20204v1 Announce Type: new Abstract: Cross-sectional stock ranking is a fundamental task in quantitative investment, relying on both temporal modeling of individual stocks and the capture o

researcharxiv-cs-lg

AdaTracker: Learning Adaptive In-Context Policy for Cross-Embodiment Active Visual Tracking

arXiv:2604.20305v1 Announce Type: new Abstract: Realizing active visual tracking with a single unified model across diverse robots is challenging, as the physical constraints and motion dynamics vary

safetyarxiv-cs-ro

AFMRL: Attribute-Enhanced Fine-Grained Multi-Modal Representation Learning in E-commerce

arXiv:2604.20135v1 Announce Type: new Abstract: Multimodal representation is crucial for E-commerce tasks such as identical product retrieval. Large representation models (e.g., VLM2Vec) demonstrate s

researcharxiv-cs-cl

ALAS: Adaptive Long-Horizon Action Synthesis via Async-pathway Stream Disentanglement

arXiv:2604.20721v1 Announce Type: new Abstract: Long-Horizon (LH) tasks in Human-Scene Interaction (HSI) are complex multi-step tasks that require continuous planning, sequential decision-making, and

researcharxiv-cs-ro

Hardware

Algorithm and Hardware Co-Design for Efficient Complex-Valued Uncertainty Estimation

arXiv:2604.19993v1 Announce Type: cross Abstract: Complex-Valued Neural Networks (CVNNs) have significant advantages in handling tasks that involve complex numbers. However, existing CVNNs are unable

hardwarearxiv-cs-lg

Are LLM Uncertainty and Correctness Encoded by the Same Features? A Functional Dissociation via Sparse Autoencoders

model-releasesarxiv-cs-cl

arXiv:2604.19974v1 Announce Type: cross Abstract: Large language models can be uncertain yet correct, or confident yet wrong, raising the question of whether their output-level uncertainty and their a

Ask Only When Needed: Proactive Retrieval from Memory and Skills for Experience-Driven Lifelong Agents

arXiv:2604.20572v1 Announce Type: new Abstract: Online lifelong learning enables agents to accumulate experience across interactions and continually improve on long-horizon tasks. However, existing me

safetyarxiv-cs-cl

Avoiding Overthinking and Underthinking: Curriculum-Aware Budget Scheduling for LLMs

arXiv:2604.19780v1 Announce Type: new Abstract: Scaling test-time compute via extended reasoning has become a key paradigm for improving the capabilities of large language models (LLMs). However, exis

safetyarxiv-cs-cl

Beyond Majority Voting: Towards Fine-grained and More Reliable Reward Signal for Test-Time Reinforcement Learning

model-releasesarxiv-cs-cl

arXiv:2512.15146v3 Announce Type: replace Abstract: Test-time reinforcement learning mitigates the reliance on annotated data by using majority voting results as pseudo-labels, emerging as a complemen

Beyond ZOH: Advanced Discretization Strategies for Vision Mamba

arXiv:2604.20606v1 Announce Type: cross Abstract: Vision Mamba, as a state space model (SSM), employs a zero-order hold (ZOH) discretization, which assumes that input signals remain constant between s

researcharxiv-cs-ai

Bimanual Robot Manipulation via Multi-Agent In-Context Learning

arXiv:2604.20348v1 Announce Type: cross Abstract: Language Models (LLMs) have emerged as powerful reasoning engines for embodied control. In particular, In-Context Learning (ICL) enables off-the-shelf

Bridging Mechanistic Interpretability and Prompt Engineering with Gradient Ascent for Interpretable Persona Control

model-releasesarxiv-cs-lg

arXiv:2601.02896v2 Announce Type: replace Abstract: Controlling emergent behavioral personas (e.g., sycophancy, hallucination) in Large Language Models (LLMs) is critical for AI safety, yet remains a

Calibrating conditional risk

arXiv:2604.20409v1 Announce Type: new Abstract: We introduce and study the problem of calibrating conditional risk, which involves estimating the expected loss of a prediction model conditional on inp

researcharxiv-cs-lg

CAST: Achieving Stable LLM-based Text Analysis for Data Analytics

arXiv:2602.15861v2 Announce Type: replace-cross Abstract: Text analysis of tabular data relies on two core operations: summarization for corpus-level theme extraction and tagging for row-level labelin

Causal-Transformer with Adaptive Mutation-Locking for Early Prediction of Acute Kidney Injury

arXiv:2604.20259v1 Announce Type: new Abstract: Accurate early prediction of Acute Kidney Injury (AKI) is critical for timely clinical intervention. However, existing deep learning models struggle wit

researcharxiv-cs-lg