AI News

Forecast Accuracy Is Not Trading Profit: Evolving Small Recurrent Networks for Stock Return Prediction (arxiv.org)

Common-Mode Errors Limit Low-Timestep Deep Spiking Q-Networks (arxiv.org)

Making COMET Comparable Across Scripts: Diagnosis and Correction of Tokeniser-Induced Script Bias in Indic MT Evaluation (arxiv.org)

CheckerBench: Can Long-Horizon Agents Synthesize Static-Analysis Checkers? (arxiv.org)

A Systematic Study of Small Language Models on Abstract Reasoning Tasks (arxiv.org)

HygieneRoboBench: Benchmarking Hygiene-Aware Planning for Household Robots (arxiv.org)

Seeing the Invisible: Physics-Guided Visual Prompting for Temperature- and Radiation-Aware VLA Navigation (arxiv.org)

TICDA: Tabular In-Context Data Attribution (arxiv.org)

OpenSplatGraph: From Dense Semantic Maps to Structured Scene Graphs for Open-Vocabulary Robot Perception (arxiv.org)

Recurrent Looped Transformer (arxiv.org)

APEX: Speculate smarter, not deeper (arxiv.org)

Learning to Retrieve via Reinforcement Learning in Embedding Space (arxiv.org)

Modeling Latent Disturbances for Robust Decision-Making in World Models (arxiv.org)

Behavioral Cloning Mystery (arxiv.org)

Calibrated Answers About Randomized Trials From a 4-Billion-Parameter Open Model: A Registered Test and a License-Clean Release (arxiv.org)

Visual-Invariance-Augmented Feature Optimal Alignment for Transferable Adversarial Attacks against Closed-Source MLLMs (arxiv.org)

EMHO: EMbodied Agent Harness Optimization via Experience Traces (arxiv.org)

Transect: Retaining Observability for Long-Horizon LLM Agent Evaluations (arxiv.org)

Explainable Failure Prediction and Prevention in Maritime (arxiv.org)

MLToolBench: Learning Tool-Augmented Agents for Machine Learning Development (arxiv.org)

Which Image Property Carries the Jailbreak? A Controlled Dissection of Image-to-Text Jailbreaks (arxiv.org)

Where Does the Audio Jailbreak Live? A Controlled Frequency-Depth Audit of AdvWave-P on Qwen2-Audio (arxiv.org)

Hierarchy-GBP: Accelerating Factor Graph Inference via Abstraction and Recovery (arxiv.org)

FactorBench: A Portfolio-Aware Benchmark for Automated Factor Mining (arxiv.org)

Extending Music Annotation Schemas: Zero-Shot Prediction or Few-Shot Adaptation? (arxiv.org)

Nonlocal Hamiltonian Dynamics on Sparse L\'evy Graphs: Spectral Analysis and Multimodal Sampling (arxiv.org)

Adaptive Power Sampling for LLM Reasoning (arxiv.org)

MINDSET: Energy-based Schema Evolution for Long Conversational Agent Memory (arxiv.org)

Recursive Game Creator: An Agentic Product-Level Experience-Oriented Game Harness (arxiv.org)

Foresight-over-Graph: Reasoning Beyond Local Horizons for Knowledge Base Question Answering (arxiv.org)

Test-Time Adaptation of Quantized ViTs via Single-Pass Quantizer-Aligned Recalibration (arxiv.org)

How Much Planning Is Enough? Reducing Search and Computation in World-Model Planning (arxiv.org)

Transferable Spatial Temporal Coherence Adversarial Attack on Black-Box Vision Language Models for Autonomous Driving (arxiv.org)

MARCO: The Radioactive Watermark for Protein Generative Models (arxiv.org)

MASC: A Multi-Agent Self-Calibration Framework with Latent Construct Alignment for Consistent Client Role-Playing in Psychological Counseling (arxiv.org)

MoF: Preference-Aware Mixture Modeling for Black-Box LLM Personalization (arxiv.org)

Calibration Is Not Control: Intervention Value for LLM-Agent Oversight (arxiv.org)

RAM-Net: Linear-Time Sequence Modeling with Sparsely Addressable State (arxiv.org)

Rethinking Fine-Tuning: Unlocking Hidden Capabilities in Vision-Language Models (arxiv.org)

Variational Physics-Informed Ansatz for Reconstructing Hidden Interaction Networks from Steady States (arxiv.org)

Optimal Transportation and Alignment Between Gaussian Measures (arxiv.org)

A Footprint-Aware, High-Resolution Approach for Carbon Flux Prediction Across Diverse Ecosystems (arxiv.org)

Micro Neural Policies for Safe Real-Time Robotic Control (arxiv.org)

Honesty over Accuracy: Trustworthy Language Models through Reinforced Hesitation (arxiv.org)

Direct Regret Optimization in Bayesian Optimization (arxiv.org)

KO: Kinetics-inspired Neural Optimizer with PDE Simulation Approaches (arxiv.org)

Causal Effect Estimation under Networked Interference without Networked Unconfoundedness Assumption (arxiv.org)

Dynamical low-rank equilibrium computation for stochastic games between advanced persistent threats and moving target defense (arxiv.org)

Variance-Optimal Off-Policy Evaluation with Conjunct Effect Modeling (arxiv.org)

Multi-Label Perceptual Bug Detection in Video Games using Deep Learning on Gameplay Footage (arxiv.org)