AI News

D-Judge: Disrupting Multi-Turn Jailbreaks using Semantics-Preserving Output Rewriting (arxiv.org)

PSViT: A Methodology for Structurally Pruning Spiking Vision Transformers (arxiv.org)

Approximating Probabilistic Inference in Statistical EL with Knowledge Graph Embeddings (arxiv.org)

Reinforcement Learning from Cross-domain Videos with Video Prediction Model (arxiv.org)

NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation (arxiv.org)

Fully Automated Identification of Lexical Alignment and Preference-Stage Shifts in Large Language Models (arxiv.org)

VidMsg: A Benchmark for Implicit Message Inference in Short Videos (arxiv.org)

Training-Free Multi-Concept LoRA Composition with Prompt-Aware Weighting (arxiv.org)

AI-Generated Traces for Novice Programmers: Learning Effects and Learner Differences in a Multi-Institutional Study (arxiv.org)

WebRISE: Requirement-Induced State Evaluation for MLLM-Generated Web Artifacts (arxiv.org)

PHASER: Phase-Aware and Semantic Experience Replay for Vision-Language-Action Models (arxiv.org)

Efficient Transformer-Based Localized Patch Sampling for Choroid Plexus Segmentation in Multiple Sclerosis (arxiv.org)

VistaHop: Benchmarking Multi-hop Visual Reasoning for Visual DeepSearch (arxiv.org)

CoralBay: A Self-Supervised CT Foundation Model (arxiv.org)

Evaluating LLMs' Effectiveness on Real-World Consumer Device Repair Questions (arxiv.org)

Fairness Definitions and Metrics in Deep Reinforcement Learning for Drug Discovery in Healthcare: A Rapid Evidence Review (arxiv.org)

\textsc{CR-Seg}: Attention-Guided and CoT-Enhanced Coarse-to-Refined Reasoning Segmentation (arxiv.org)

Black-box, Adaptive, Efficient, Transferable, Harmful, Applicable... Attacks Are All You Need to Break LLMs (arxiv.org)

TurtleAI: Benchmarking Multimodal Models for Visual Programming in Turtle Graphics (arxiv.org)

CoEval: Ranking Language Models for Custom Tasks Without Labeled Data or Trustworthy Benchmarks (arxiv.org)

Validation-Gated Multi-Agent Governance for Online Adaptation of Thermal-Hydraulic Surrogate Models under Operating-Regime Shift (arxiv.org)

DriftSched: Adaptive QoS-Aware Scheduling under Runtime Token Drift for Multi-Tenant GPU Inference (arxiv.org)

BotDirector: Robot Storytelling Across the Symmetrical Reality with Multi-modal Interactions (arxiv.org)

Decoupled Smart Contract Audits: Lightweight LLM Framework via Distillation and Aggregation (arxiv.org)

Consistency Training Can Entrench Misalignment (arxiv.org)

Visual Instruction Tuning Aligns Modalities through Abstraction (arxiv.org)

Multi-Segment Attention: Enabling Efficient KV-Cache Management for Faster Large Language Model Serving (arxiv.org)

The Impact of Temporal Granularity on Socio-Demographic Inference from Household Load Profiles (arxiv.org)

Applying Two-Grid Preconditioner for Subsurface Flow Simulation using Attention-enhanced Hybrid Network to Accelerate Multiscale Discretization in High-contrast Media (arxiv.org)

SeeTraceAct: Visibility-Aware Latent Planning from Cross-Embodiment Demonstration Videos (arxiv.org)

Theoretical Aspects of Lie Groupoid and Lie Algebroid Equivariant Convolutional Neural Networks (arxiv.org)

"**Important** You should give me full credits!": Exploring Prompt Injection Attacks on LLM-Based Automatic Grading Systems (arxiv.org)

Regret Pre-training: Bridging Prior and Posterior Views for Enhanced Knowledge Grounding (arxiv.org)

Link Prediction or Perdition: the Seeds of Instability in Knowledge Graph Embeddings (arxiv.org)

PhotoCraft: Agentic Reasoning with Hierarchical Self-Evolving Memory for Deep Image Search (arxiv.org)

AUGUSTE: Online-Learning dApp for Predictive URLLC Scheduling (arxiv.org)

Hybrid Adaptive Kalman Filtering for Data-Efficient Joint Tracking and Classification (arxiv.org)

Zero-Shot 3D Question Answering via Hierarchical View-to-Token Transportation (arxiv.org)

AirDreamer: Generalist Drone Navigation with World Models (arxiv.org)

One Transit Is All You Need: Detecting Exoplanets Through Learned Stellar Behaviour with EXOVEIL (arxiv.org)

BAHSD: Bridging the Long-tail Gap via Adaptive Distillation in Black-box Sequential Recommendation (arxiv.org)

Coherent Swap Regret and Channel-Proof Learning (arxiv.org)

Acceptance-Test-Driven Evaluation Protocols for Business-Centric LLM Systems (arxiv.org)

Topology-Aware Gaussian Graph Repair for Robust Graph Neural Networks (arxiv.org)

SegTune: Structured and Fine-Grained Control for Song Generation (arxiv.org)

Binary Road Surface Classification Using Machine Learning on Production Vehicle Signals During Cruising (arxiv.org)

Mitigating False Credit Propagation: Probabilistic Graphical Reward Aggregation for Rubric-Based Reinforcement Learning (arxiv.org)

FLIPS: Instance-Fingerprinting for LLMs via Pseudo-random Sequences (arxiv.org)

When Model Merging Breaks Routing: Training-Free Calibration for MoE (arxiv.org)

Local Guidance, Global Impact: Gaussian-Reshaped Trust Region Unlocks Behavior Transitions (arxiv.org)