The Validity Gap in Health AI Evaluation: A Cross-Sectional Analysis of Benchmark Composition
arXiv:2603.18294v1 Announce Type: new Abstract: Background: Clinical trials rely on transparent inclusion criteria to ensure generalizability. In contrast, benchmarks validating health-related large language models (LLMs) rarely characterize the "patient" or "query" populations they contain. Without defined composition, aggregate performance metrics...
Cross-Domain Demo-to-Code via Neurosymbolic Counterfactual Reasoning
arXiv:2603.18495v1 Announce Type: new Abstract: Recent advances in Vision-Language Models (VLMs) have enabled video-instructed robotic programming, allowing agents to interpret video demonstrations and generate executable control code. We formulate video-instructed robotic programming as a cross-domain adaptation problem, where perceptual and...
TherapyGym: Evaluating and Aligning Clinical Fidelity and Safety in Therapy Chatbots
arXiv:2603.18008v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for mental-health support; yet prevailing evaluation methods--fluency metrics, preference tests, and generic dialogue benchmarks--fail to capture the clinically critical dimensions of psychotherapy. We introduce THERAPYGYM, a framework that...
Reasonably reasoning AI agents can avoid game-theoretic failures in zero-shot, provably
arXiv:2603.18563v1 Announce Type: new Abstract: AI agents are increasingly deployed in interactive economic environments characterized by repeated AI-AI interactions. Despite AI agents' advanced capabilities, empirical studies reveal that such interactions often fail to stably induce a strategic equilibrium, such as...
Learned but Not Expressed: Capability-Expression Dissociation in Large Language Models
arXiv:2603.18013v1 Announce Type: new Abstract: Large language models (LLMs) demonstrate the capacity to reconstruct and trace learned content from their training data under specific elicitation conditions, yet this capability does not manifest in standard generation contexts. This empirical observational study...
Controllable Evidence Selection in Retrieval-Augmented Question Answering via Deterministic Utility Gating
arXiv:2603.18011v1 Announce Type: new Abstract: Many modern AI question-answering systems convert text into vectors and retrieve the closest matches to a user question. While effective for topical similarity, similarity scores alone do not explain why some retrieved text can serve...
How LLMs Distort Our Written Language
arXiv:2603.18161v1 Announce Type: new Abstract: Large language models (LLMs) are used by over a billion people globally, most often to assist with writing. In this work, we demonstrate that LLMs not only alter the voice and tone of human writing,...
GRAFITE: Generative Regression Analysis Framework for Issue Tracking and Evaluation
arXiv:2603.18173v1 Announce Type: new Abstract: Large language models (LLMs) are largely motivated by their performance on popular topics and benchmarks at the time of their release. However, over time, contamination occurs due to significant exposure of benchmark data during training....
From Noise to Signal: When Outliers Seed New Topics
arXiv:2603.18358v1 Announce Type: new Abstract: Outliers in dynamic topic modeling are typically treated as noise, yet we show that some can serve as early signals of emerging topics. We introduce a temporal taxonomy of news-document trajectories that defines how documents...
TopoChunker: Topology-Aware Agentic Document Chunking Framework
arXiv:2603.18409v1 Announce Type: new Abstract: Current document chunking methods for Retrieval-Augmented Generation (RAG) typically linearize text. This forced linearization strips away intrinsic topological hierarchies, creating ``semantic fragmentation'' that degrades downstream retrieval quality. In this paper, we propose TopoChunker, an agentic...
WASD: Locating Critical Neurons as Sufficient Conditions for Explaining and Controlling LLM Behavior
arXiv:2603.18474v1 Announce Type: new Abstract: Precise behavioral control of large language models (LLMs) is critical for complex applications. However, existing methods often incur high training costs, lack natural language controllability, or compromise semantic coherence. To bridge this gap, we propose...
EntropyCache: Decoded Token Entropy Guided KV Caching for Diffusion Language Models
arXiv:2603.18489v1 Announce Type: new Abstract: Diffusion-based large language models (dLLMs) rely on bidirectional attention, which prevents lossless KV caching and requires a full forward pass at every denoising step. Existing approximate KV caching methods reduce this cost by selectively updating...
When Names Change Verdicts: Intervention Consistency Reveals Systematic Bias in LLM Decision-Making
arXiv:2603.18530v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for high-stakes decisions, yet their susceptibility to spurious features remains poorly characterized. We introduce ICE-Guard, a framework applying intervention consistency testing to detect three types of spurious feature...
Implicit Grading Bias in Large Language Models: How Writing Style Affects Automated Assessment Across Math, Programming, and Essay Tasks
arXiv:2603.18765v1 Announce Type: new Abstract: As large language models (LLMs) are increasingly deployed as automated graders in educational settings, concerns about fairness and bias in their evaluations have become critical. This study investigates whether LLMs exhibit implicit grading bias based...
Mi:dm K 2.5 Pro
arXiv:2603.18788v1 Announce Type: new Abstract: The evolving LLM landscape requires capabilities beyond simple text generation, prioritizing multi-step reasoning, long-context understanding, and agentic workflows. This shift challenges existing models in enterprise environments, especially in Korean-language and domain-specific scenarios where scaling is...
Engineering Verifiable Modularity in Transformers via Per-Layer Supervision
arXiv:2603.18029v1 Announce Type: new Abstract: Transformers resist surgical control. Ablating an attention head identified as critical for capitalization produces minimal behavioral change because distributed redundancy compensates for damage. This Hydra effect renders interpretability illusory: we may identify components through correlation,...
InfoMamba: An Attention-Free Hybrid Mamba-Transformer Model
arXiv:2603.18031v1 Announce Type: new Abstract: Balancing fine-grained local modeling with long-range dependency capture under computational constraints remains a central challenge in sequence modeling. While Transformers provide strong token mixing, they suffer from quadratic complexity, whereas Mamba-style selective state-space models (SSMs)...
Fundamental Limits of Neural Network Sparsification: Evidence from Catastrophic Interpretability Collapse
arXiv:2603.18056v1 Announce Type: new Abstract: Extreme neural network sparsification (90% activation reduction) presents a critical challenge for mechanistic interpretability: understanding whether interpretable features survive aggressive compression. This work investigates feature survival under severe capacity constraints in hybrid Variational Autoencoder--Sparse Autoencoder...
Variational Phasor Circuits for Phase-Native Brain-Computer Interface Classification
arXiv:2603.18078v1 Announce Type: new Abstract: We present the \textbf{Variational Phasor Circuit (VPC)}, a deterministic classical learning architecture operating on the continuous $S^1$ unit circle manifold. Inspired by variational quantum circuits, VPC replaces dense real-valued weight matrices with trainable phase shifts,...
SLEA-RL: Step-Level Experience Augmented Reinforcement Learning for Multi-Turn Agentic Training
arXiv:2603.18079v1 Announce Type: new Abstract: Large Language Model (LLM) agents have shown strong results on multi-turn tool-use tasks, yet they operate in isolation during training, failing to leverage experiences accumulated across episodes. Existing experience-augmented methods address this by organizing trajectories...
MolRGen: A Training and Evaluation Setting for De Novo Molecular Generation with Reasonning Models
arXiv:2603.18256v1 Announce Type: new Abstract: Recent advances in reasoning-based large language models (LLMs) have demonstrated substantial improvements in complex problem-solving tasks. Motivated by these advances, several works have explored the application of reasoning LLMs to drug discovery and molecular design....
Discovering What You Can Control: Interventional Boundary Discovery for Reinforcement Learning
arXiv:2603.18257v1 Announce Type: new Abstract: Selecting relevant state dimensions in the presence of confounded distractors is a causal identification problem: observational statistics alone cannot reliably distinguish dimensions that correlate with actions from those that actions cause. We formalize this as...
Sharpness-Aware Minimization in Logit Space Efficiently Enhances Direct Preference Optimization
arXiv:2603.18258v1 Announce Type: new Abstract: Direct Preference Optimization (DPO) has emerged as a popular algorithm for aligning pretrained large language models with human preferences, owing to its simplicity and training stability. However, DPO suffers from the recently identified squeezing effect...
Detection Is Cheap, Routing Is Learned: Why Refusal-Based Alignment Evaluation Fails
arXiv:2603.18280v1 Announce Type: new Abstract: Current alignment evaluation mostly measures whether models encode dangerous concepts and whether they refuse harmful requests. Both miss the layer where alignment often operates: routing from concept detection to behavioral policy. We study political censorship...
Path-Constrained Mixture-of-Experts
arXiv:2603.18297v1 Announce Type: new Abstract: Sparse Mixture-of-Experts (MoE) architectures enable efficient scaling by activating only a subset of parameters for each input. However, conventional MoE routing selects each layer's experts independently, creating N^L possible expert paths -- for N experts...
A Family of Adaptive Activation Functions for Mitigating Failure Modes in Physics-Informed Neural Networks
arXiv:2603.18328v1 Announce Type: new Abstract: Physics-Informed Neural Networks(PINNs) are a powerful and flexible learning framework that has gained significant attention in recent years. It has demonstrated strong performance across a wide range of scientific and engineering problems. In parallel, wavelets...
Seeking Universal Shot Language Understanding Solutions
arXiv:2603.18448v1 Announce Type: new Abstract: Shot language understanding (SLU) is crucial for cinematic analysis but remains challenging due to its diverse cinematographic dimensions and subjective expert judgment. While vision-language models (VLMs) have shown strong ability in general visual understanding, recent...
AIMER: Calibration-Free Task-Agnostic MoE Pruning
arXiv:2603.18492v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) language models increase parameter capacity without proportional per-token compute, but the deployment still requires storing all experts, making expert pruning important for reducing memory and serving overhead. Existing task-agnostic expert pruning methods are...
Volume 2026, No. 1 – Wisconsin Law Review – UW–Madison
Contract Law and Civil Justice in Local Courts by Cathy Hwang & Justin Weinstein-Tull; Preempting Drug Price Reform by Shweta Kumar; Lessons Learned? COVID’s Continued Impact on Remote Work Disability Accommodations by D’Andra Millsap Shu; Unbundling AI Openness by Parth...
Catching Pokémon, Not Tax Bills
Introduction What if we told you that you could play a unique and magical game for free? What if we told you this game would let you chase fantastical creatures across your neighborhood, turning your daily stroll into an epic...