AI News

S-GAP: Stealthy Graph Attack Primitives for Graph Large Language Models (arxiv.org)

LocalIntent: Outcome-Specific Structured Generation for Backend-Compatible Smart Building Automation with Local LLMs (arxiv.org)

Trapped in Your Story: Evaluating Narrative-Induced Sycophancy in Large Language Models (arxiv.org)

SkillBrowser: Selective Skill Exposure for Scalable Skill-Using Agents (arxiv.org)

Darwin-Claw: Environment-Adaptive Experience Consolidation for Recursively Self-Evolving Agents (arxiv.org)

RL-Guided Temporal Localization for Dual-Channel Retrieval in Long-Horizon Agent Memory (arxiv.org)

From Trajectories to Differentiated Skills: Cognitive Differentiation for Self-Improving Agents (arxiv.org)

GEMS: Molecular Structure Elucidation as Learned Search over Isomer Graphs (arxiv.org)

OwlSight: A Panoramic Monitor for Chain-of-Thought Reasoning (arxiv.org)

Beyond Large Language Models: A Mental World Model Framework for Mental Modeling, Simulation, Evolution, and Interpretability (arxiv.org)

DrugHarness: An Evolvable Agent Harness for Computational Drug Discovery (arxiv.org)

FuTransformer: FP8 prefill and INT4 decode for Llama-style inference on Hopper (arxiv.org)

Beyond SAEs: Consistent Concept Discovery with Sparse Coding (arxiv.org)

Self-Building Labs (arxiv.org)

AI-Driven Reproducibility and Analytical Robustness in Scientific Discovery (arxiv.org)

Eliciting Specifications: When Does Interactive Clarification Help? (arxiv.org)

Towards Adaptive and Real-Survey-Aligned Simulated Urban Mobility Behaviour of Generative Agents (arxiv.org)

From Individual Songs to Music Corpora: Evaluating Explainable Harmonic Decision-Making on HookTheory (arxiv.org)

Really Need Recursion? Revisiting One-pass Registration with Multi-level Multi-space Contextual Guidance (arxiv.org)

Measuring Mechanistic Uncertainty in Large Language Models via Causal Robustness (arxiv.org)

Recoverability-Controlled Safety Miscalibration in Large Language Models (arxiv.org)

DriveWorld3D Evidence Constrained 3D World Modeling and Generative Simulation for Real-World Autonomous Driving (arxiv.org)

Political Narratives in the 2024 Moldovan Presidential Election: Stance, Amplification, and Influence (arxiv.org)

AuditKIT: A Unified Evaluation Spine for Benchmarking, Judging, and Comparing Large Language Models and Agentic Systems (arxiv.org)

Harnessing Trust in Directed Graphs: Redefining Robustness of Graph Learning (arxiv.org)

Mitigating Demographic Disparities in Explanation-Human Alignment via Rationale Regularization (arxiv.org)

TALLOP: Joint Optimization of Local Training and Communication Compression for Time-Constrained Federated Learning (arxiv.org)

StratFL: Countering Covert Model Poisoning Attacks with Strategic Federated Learning (arxiv.org)

A Label Noise Robust Algorithm for Graph Data using Neighborhood Label Distribution (arxiv.org)

A Plausible Explanation Is Enough to Mislead Multimodal Judges (arxiv.org)

A Minimal ReLU Classifier on Non-Linearly Separable Data: Convergence and Implicit Bias at and beyond the Edge of Stability (arxiv.org)

"Deep lifts": Neural networks can generalize even with the dumbest optimizer (arxiv.org)

Towards Looped Models Done Right, Part I: Topology, Input Injection, Recurrent-State Design (arxiv.org)

GR-OPD: Group-Relative Sequence Credit for On-Policy Distillation (arxiv.org)

Path to Prior-Data Fitted Relational Models via Learned Latent Multi-table Representations (arxiv.org)

Where Latents Belong and How They Flow: Joint Identification of Structural Roles (arxiv.org)

Towards a GraphNARX model for structural dynamics (arxiv.org)

Domain-specific language alignment of multimodal embeddings towards explainable predictions (arxiv.org)

LED: Improving Masked Diffusion Language Models by Lookahead Editing (arxiv.org)

Pluvio: Assembly Clone Search for Out-of-domain Architectures and Libraries through Transfer Learning and Conditional Variational Information Bottleneck (arxiv.org)

Beyond Boundaries: A Comprehensive Survey of Transferable Attacks on AI Systems (arxiv.org)

VietMed: A Dataset and Benchmark for Automatic Speech Recognition of Vietnamese in the Medical Domain (arxiv.org)

Harmful YouTube Video Detection: A Taxonomy of Online Harm and MLLMs as Alternative Annotators (arxiv.org)

Paint Outside the Box: Synthesizing and Selecting Training Data for Visual Grounding (arxiv.org)

MetaCurate: Evolving Curation Filters for High-Quality Agentic Coding Data (arxiv.org)

Local proxies for memory-efficient training (arxiv.org)

Grounded Credit Assignment in Reinforcement Learning for Multi-Turn Search Agents (arxiv.org)

Nearly Linear Private Learning of Littlestone Classes (arxiv.org)

Non-Parametric Optimization for Scalable Learning in Stochastic Decision Problems (arxiv.org)

Rethinking Negative Prompting for Concept Suppression: Unconditional Anchoring and Localized Guidance (arxiv.org)