AI News

HERO: Hindsight-Enhanced Reflection from Environment Observations for Agentic Self-Distillation (arxiv.org)

TouchThinker: Scaling Tactile Commonsense Reasoning to the Open World with Large-scale Data and Action-aware Representation (arxiv.org)

TreeSeeker: Tree-Structured Trial, Error, and Return in Deep Search (arxiv.org)

Lung-R1: A Knowledge Graph-Guided LLM for Pulmonary Diagnostic Reasoning (arxiv.org)

Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents (arxiv.org)

Mind the Perspective: Let's Reason Recursively for Theory of Mind (arxiv.org)

When Do Data-Driven Systems Exhibit the Capability to Infer? (arxiv.org)

SVoT: State-aware Visualization-of-Thought for Spatial Reasoning via Reinforcement Learning (arxiv.org)

Toward Trustworthy AI: Multi-Target Adversarial Attacks and Robust Defenses for Continuous Data Summarization (arxiv.org)

Embodied-BenchClaw: An Autonomous Multi-Agent System for Embodied Spatial Intelligence Benchmark Construction (arxiv.org)

The Art of Interrogation: Consistency Amplifies Factuality in Spatial Reasoning (arxiv.org)

MODF-SIR: A Multi-agent Omni-modal Distilled Framework for Social Intelligence Reasoning (arxiv.org)

Human-Enhanced Loop Modeling (HELM): Agent-Based Finite Element Modeling of Concrete Bridge Barriers (arxiv.org)

Existential Indifference: Self-Nonpreservation as a Necessary Architectural Condition for Aligned Superintelligence (or: The Suicidal AI) (arxiv.org)

A Lightweight Multi-Agent Framework for Automated Concrete Barrier Design (arxiv.org)

Automating Geometry-Intensive Compliance Checking in BIM: Graph-Based Semantic Reasoning Framework (arxiv.org)

IntElicit: Eliciting and Assessing Contextualized Creativity via Dialogue Policy Optimization (arxiv.org)

Towards Responsibly Non-Compliant Machines (arxiv.org)

The Impossibility of Eliciting Latent Knowledge (arxiv.org)

A Five-Plane Reference Architecture for Runtime Governance of Production AI Agents (arxiv.org)

PROJECTMEM: A Local-First, Event-Sourced Memory and Judgment Layer for AI Coding Agents (arxiv.org)

Nonslop: A Gamified Experiment in Human-AI Collaborative Writing (arxiv.org)

From Architecture to Output: Structural Origins of Hallucination in Large Language Models and the Amplifying Role of Data (arxiv.org)

PoQ-Judge: A Multi-Architecture Evaluation Framework for Cost-Aware Proof-of-Quality in Decentralized LLM Inference (arxiv.org)

MA-DLE: Speech-based Automatic Depression Level Estimation via Memory Augmentation (arxiv.org)

The Structural Attention Tax: How Retrieval Format Hijacks In-Context Learning Independent of Content (arxiv.org)

NightFeats @ MMU-RAGent NeurIPS 2025: A Context-Optimized Multi-Agent RAG System for the Text-to-Text Track (arxiv.org)

Hybrid Iterative Neural Low-Regularity Integrator for Nonlinear Dispersive Equations (arxiv.org)

RePAIR: Predictive Self-Supervised Representation Learning in Chess (arxiv.org)

Attention by Synchronization in Coupled Oscillator Networks (arxiv.org)

MemNovo: Look Back at the Spectrum for Balanced De Novo Peptide Sequencing from Mass Spectrometry (arxiv.org)

Beyond representational alignment with brain-guided language models for robust reasoning (arxiv.org)

HAMNO: A Hierarchical Adaptive Multi-scale Neural Operator with Physics-Informed Learning for Dynamical Systems (arxiv.org)

PAWS: Preference Learning with Advantage-Weighted Segments (arxiv.org)

Time-Series Foundation Model Embeddings for Remaining Useful Life Estimation (arxiv.org)

Tabular Foundation Models for Clinical Survival Analysis via Survival-Aware Adaptation (arxiv.org)

Generalization Hacking: Models Can Game Reinforcement Learning by Preventing Behavioral Generalization (arxiv.org)

Reliable Error Estimation for PINNs: Lower and Upper A Posteriori Bounds (arxiv.org)

Simplicity Suffices for Parameter Noise Injection in Stochastic Gradient Descent (arxiv.org)

Neural-Parameterized Cellular Automata for Wildfire Spread (arxiv.org)

Efficient Time Series Clustering from Multiscale Reservoir Dynamics with Granular-Ball Anchoring Graph Optimization (arxiv.org)

A Riemannian Approach to Low-Rank Optimal Transport (arxiv.org)

Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders (arxiv.org)

SpAArSIST: Sparsified AASIST for Efficient and Reliable Anti-Spoofing (arxiv.org)

Latent World Recovery for Multimodal Learning with Missing Modalities (arxiv.org)

PCA-Enhanced Adaptive NVAR Framework for High-Resolution Sea Surface Temperature Forecasting in the East Sea (arxiv.org)

nD-RoPE: A Generalized RoPE for n-Dimensional Position Embedding (arxiv.org)

How Low Can You Go? Active Learning for Sparse Model Discovery in the Ultra-Low-Data Limit (arxiv.org)

Implicit Neural Representations of Individual Behavior (arxiv.org)

Re-evaluating Confidence Remasking in Masked Diffusion Language Models (arxiv.org)