AI News

DualStake: Dual-Path Confidence Calibration in Deep Research Agents (arxiv.org)

mzCache: On-Device LLM Memory Management under Multitasking (arxiv.org)

Right Frame, Wrong Rule: Cultural Cues Expose the Financial Knowledge Gap They Were Meant to Close (arxiv.org)

SinkPruner: Sink-Free Visual Token Pruning for Multimodal Large Language Models (arxiv.org)

Web Price Extraction: State of the Art and an Adaptive Browserless Implementation (arxiv.org)

Probabilistic Multi-Agent Aircraft Landing Time Prediction (arxiv.org)

CADKnitter: Compositional CAD Generation from Text and Geometry Guidance (arxiv.org)

Auditing Frozen-Encoder Anomaly Detection Across Mechanical Systems: Representation Provenance, Calibration, and Protocol Effects (arxiv.org)

Online Regime-aware Calibration for Black-box Social Simulators via Posterior-assisted Evolutionary Dynamic Optimization (arxiv.org)

Is Knowledge Distillation Actually Greener? A Case Study in Machine Translation (arxiv.org)

RAPIDMap: Rapid Multi-Agent Pipeline for Interpretable Disaster Mapping from Satellite and Street-view Imagery (arxiv.org)

GUI-CC: Benchmarking Contextual Consistency of GUI World Models as Agent Environments (arxiv.org)

From Detection to Refusal: Safer LLMs via Circuit-Guided Weight Scaling (arxiv.org)

Zero-Shot Respiratory Sound Classification through LLM-Augmented Audio-Text Alignment (arxiv.org)

Do Multimodal LLMs See Before They Read? Diagnosing Contextual Sycophancy (arxiv.org)

Life Operators: a self-evolving framework for multiscale life modelling (arxiv.org)

Auditing Harness Tampering in Self-Improving Agents (arxiv.org)

AutoXRD: Autonomous LLM Agents and Comprehensive Evaluation for Powder Diffraction Analysis (arxiv.org)

Intelligent Edge Computing (arxiv.org)

KItCAT: Knowledge Injection via Input Corruption for Auto-regressive Training (arxiv.org)

EDRAC: Benchmarking Arabic Dialect Reading Comprehension (arxiv.org)

Retrieval, Scoring, and Decoding Shape Performance and Stability in LLM-based Conversational Recommendation (arxiv.org)

Commit-first LLM judging inherits the judge's own errors (arxiv.org)

RecalibrateGPT: AI Fatigue Resilient Conversational Interfaces (arxiv.org)

The Interlingua Hypothesis: LLMs Translate via a Latent Task-agnostic Feature Space (arxiv.org)

Predicting Program Exit Code with LLMs and Programming Language Semantics (arxiv.org)

SoK: When Safe Agents Fail Together: The Security of Multi Agent LLM Systems (arxiv.org)

Towards Effective Structured Context Modeling for Conversational Recommender Systems via Dual-node Monte Carlo Tree Search (arxiv.org)

TUTTI: Toward generalizable audio-to-score transcription via fully synthesized data (arxiv.org)

Visual Framing for News Stance Detection via Image Generation (arxiv.org)

From Terminology to Diagrams: Visual-Instruction Generation for Scientific Diagram Understanding (arxiv.org)

Calibration is the Bottleneck: An Action-Class Diagnostic of Multi-Turn Tool-Calling (arxiv.org)

The zbMATH Open Knowledge Graph: Tracing Centuries of Mathematical Research (arxiv.org)

Disclosure-Gated User Simulation for Companion-Agent Evaluation (arxiv.org)

Semi-Supervised Virtual Staining via Morphology Preservation and Histopathological Realism Constraints (arxiv.org)

On the Human and Computer Alignment of Attribute-Based Music Matches (arxiv.org)

Inspicio: Open-Vocabulary, LLM-Based Sense Retrieval for Historical Languages (arxiv.org)

A Network Science Perspective on Evaluating Deep Graph Generative Models (arxiv.org)

On Synthesis of Metric Interval Temporal Logics (arxiv.org)

Causal Evidentiary Governance for High-Risk Machine Learning Systems (arxiv.org)

FedReview: Review and Dispose Poisoned Updates without Validation Datasets or Historic Knowledge (arxiv.org)

HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation (arxiv.org)

Lagged Coupling: Internal Representations Become Readable Before They Become Causal (arxiv.org)

trajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories (arxiv.org)

ToSCA: Leveraging Hierarchical Reinforcement Learning on Temporal and Strategic Abstractions of Conversational Agents (arxiv.org)

MRMAD: A Multi-Round Multi-Audio Benchmark for Evaluating Acoustic Degradation Perception in Large Audio-Language Models (arxiv.org)

Freeze, Diffuse, Decode: Geometry-Aware Adaptation of Pretrained Transformer Embeddings for Antimicrobial Peptide Design (arxiv.org)

Scaling Near-Optimal SFT-RL Annotation Budget Allocation from Small to Large LLMs (arxiv.org)

Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulation (arxiv.org)

Beyond Scores: Understanding LLM-as-a-Judge Mechanisms in Summarization Evaluation (arxiv.org)