AI Research News Feeds for December 31st, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• Pre-review to Peer review: Pitfalls of Automating Reviews using Large Language Models
• Rethinking Leveraging Pre-Trained Multi-Layer Representations for Speaker Verification
• Adaptive GPU Resource Allocation for Multi-Agent Collaborative Reasoning in Serverless Environments
• Practical challenges of control monitoring in frontier AI deployments
• Solving Multi-Agent Multi-Goal Path Finding Problems in Polynomial Time
• BitFlipScope: Scalable Fault Localization and Recovery for Bit-Flip Corruptions in LLMs
• iOS as Acceleration
• Interpretable Link Prediction in AI-Driven Cancer Research: Uncovering Co-Authorship Patterns
• Literature Mining System for Nutraceutical Biosynthesis: From AI Framework to Biological Insight
• Scalable Cloud-Native Architectures for Intelligent PMU Data Processing
• Agentic Software Issue Resolution with Large Language Models: A Survey
• ReVEAL: GNN-Guided Reverse Engineering for Formal Verification of Optimized Multipliers
• Beyond Single Bugs: Benchmarking Large Language Models for Multi-Vulnerability Detection
• Cost-Aware Text-to-SQL: An Empirical Study of Cloud Compute Costs for LLM-Generated Queries
• AI-Generated Code Is Not Reproducible (Yet): An Empirical Study of Dependency Gaps in LLM-Based Coding Agents
• Space AI: Leveraging Artificial Intelligence for Space to Improve Life on Earth
• Efficient Multi-Model Orchestration for Self-Hosted Large Language Models
• A Unified AI, Embedded, Simulation, and Mechanical Design Approach to an Autonomous Delivery Robot
• Emergence of Human to Robot Transfer in Vision-Language-Action Models
• Nightjar: Dynamic Adaptive Speculative Decoding for Large Language Models Serving
• Hierarchical Pedagogical Oversight: A Multi-Agent Adversarial Framework for Reliable AI Tutoring
• Chord Recognition with Deep Learning
• Robust LLM-based Column Type Annotation via Prompt Augmentation with LoRA Tuning
• Reach-Avoid Differential game with Reachability Analysis for UAVs: A decomposition approach
• FasterPy: An LLM-based Code Execution Efficiency Optimization Framework
• The body is not there to compute: Comment on "Informational embodiment: Computational role of information structure in codes and robots" by Pitti et al
• Agentic AI for Cyber Resilience: A New Security Paradigm and Its System-Theoretic Foundations
• DECEPTICON: How Dark Patterns Manipulate Web Agents
• Heterogeneity in Multi-Agent Reinforcement Learning
• Viability and Performance of a Private LLM Server for SMBs: A Benchmark Analysis of Qwen3-30B on Consumer-Grade Hardware
• It's a TRAP! Task-Redirecting Agent Persuasion Benchmark for Web Agents
• Constraint programming model and biased random-key genetic algorithm for the single-machine coupled task scheduling problem with exact delays to minimize the makespan
• EquaCode: A Multi-Strategy Jailbreak Approach for Large Language Models via Equation Solving and Code Completion
• Explainable Neural Inverse Kinematics for Obstacle-Aware Robotic Manipulation: A Comparative Analysis of IKNet Variants
• AGRO-SQL: Agentic Group-Relative Optimization with High-Fidelity Data Synthesis
• Securing the AI Supply Chain: What Can We Learn From Developer-Reported Security Issues and Solutions of AI Projects?
• PINNs for Electromagnetic Wave Propagation
• Mobile-Efficient Speech Emotion Recognition Using DistilHuBERT: A Cross-Corpus Validation Study
• Agentic AI for Autonomous Defense in Software Supply Chain Security: Beyond Provenance to Vulnerability Mitigation
• Theory of Mind for Explainable Human-Robot Interaction
• Act2Goal: From World Model To General Goal-conditioned Policy
• Toward Trustworthy Agentic AI: A Multimodal Framework for Preventing Prompt Injection Attacks
• TPTU: Large Language Model-based AI Agents for Task Planning and Tool Usage
• Learnable WSN Deployment of Evidential Collaborative Sensing Model
• Multi-agent Self-triage System with Medical Flowcharts
• Never-Ending Behavior-Cloning Agent for Robotic Manipulation
• Bidirectional RAG: Safe Self-Improving Retrieval-Augmented Generation Through Multi-Stage Validation
• Emergent Persuasion: Will LLMs Persuade Without Being Prompted?
• GamiBench: Evaluating Spatial Reasoning and 2D-to-3D Planning Capabilities of MLLMs with Origami Folding Tasks
• Toward Equitable Recovery: A Fairness-Aware AI Framework for Prioritizing Post-Flood Aid in Bangladesh
• With Great Capabilities Come Great Responsibilities: Introducing the Agentic Risk & Capability Framework for Governing Agentic AI Systems
• Subgoaling Relaxation-based Heuristics for Numeric Planning with Infinite Actions
• HalluMat: Detecting Hallucinations in LLM-Generated Materials Science Content Through Multi-Stage Verification
• Lightweight Inference-Time Personalization for Frozen Knowledge Graph Embeddings
• DarkPatterns-LLM: A Multi-Layer Benchmark for Detecting Manipulative and Harmful AI Behavior
• Multi-AI Agent Framework Reveals the "Oxide Gatekeeper" in Aluminum Nanoparticle Oxidation
• Lessons from Neuroscience for AI: How integrating Actions, Compositional Structure and Episodic Memory could enable Safe, Interpretable and Human-Like AI
• SANet: A Semantic-aware Agentic AI Networking Framework for Cross-layer Optimization in 6G
• Tyee: A Unified, Modular, and Fully-Integrated Configurable Toolkit for Intelligent Physiological Health Care
• LLM Agents as VC investors: Predicting Startup Success via RolePlay-Based Collective Simulation
• The Wisdom of Deliberating AI Crowds: Does Deliberation Improve LLM-Based Forecasting?
• DICE: Discrete Interpretable Comparative Evaluation with Probabilistic Scoring for Retrieval-Augmented Generation
• TravelBench: A Real-World Benchmark for Multi-Turn and Tool-Augmented Travel Planning
• SAMP-HDRL: Segmented Allocation with Momentum-Adjusted Utility for Multi-agent Portfolio Management via Hierarchical Deep Reinforcement Learning
• Problems With Large Language Models for Learner Modelling: Why LLMs Alone Fall Short for Responsible Tutoring in K--12 Education
• Why We Need a New Framework for Emotional Intelligence in AI
• From Model Choice to Model Belief: Establishing a New Measure for LLM-Based Research
• TCEval: Using Thermal Comfort to Assess Cognitive and Perceptual Abilities of AI
• On Conformant Planning and Model-Checking of $\exists^*\forall^*$ Hyperproperties
• MindWatcher: Toward Smarter Multimodal Tool-Integrated Reasoning
• The World Is Bigger! A Computationally-Embedded Perspective on the Big World Hypothesis
• The Gaining Paths to Investment Success: Information-Driven LLM Graph Reasoning for Venture Capital Prediction
• Why AI Safety Requires Uncertainty, Incomplete Preferences, and Non-Archimedean Utilities
• Divergent-Convergent Thinking in Large Language Models for Creative Problem Generation
• Physics-Informed Neural Networks for Device and Circuit Modeling: A Case Study of NeuroSPICE
• GPU-Virt-Bench: A Comprehensive Benchmarking Framework for Software-Based GPU Virtualization Systems
• SoDA: An Efficient Interaction Paradigm for the Agentic Web
• HLS4PC: A Parametrizable Framework For Accelerating Point-Based 3D Point Cloud Models on FPGA