Training In-Context and In-Weights Mixtures Via Contrastive Context Sampling
arXiv:2604.01601v1 Announce Type: new Abstract: We investigate training strategies that co-develop in-context learning (ICL) and in-weights learning (IWL), and the ability to switch between them based on context relevance. Although current LLMs exhibit both modes, standard task-specific fine-tuning often erodes...
Prominent Scientists, Faith Leaders, Policymakers and Artists Call for a Prohibition on Superintelligence, as Poll Shows Americans Don’t Want It
Initial signatories include AI pioneers Yoshua Bengio and Geoffrey Hinton, leading media voices Steve Bannon and Glenn Beck, Obama's National Security Advisor Susan Rice, business trailblazers Steve Wozniak and Richard Branson, five Nobel Laureates, former Irish President Mary Robinson, actors...
The Compression Paradox in LLM Inference: Provider-Dependent Energy Effects of Prompt Compression
arXiv:2603.23528v1 Announce Type: new Abstract: The rapid proliferation of Large Language Models has created an environmental paradox: the very technology that could help solve climate challenges is itself becoming a significant contributor to global carbon emissions. We test whether prompt...
MSA: Memory Sparse Attention for Efficient End-to-End Memory Model Scaling to 100M Tokens
arXiv:2603.23516v1 Announce Type: new Abstract: Long-term memory is a cornerstone of human intelligence. Enabling AI to process lifetime-scale information remains a long-standing pursuit in the field. Due to the constraints of full-attention architectures, the effective context length of large language...
Visuospatial Perspective Taking in Multimodal Language Models
arXiv:2603.23510v1 Announce Type: new Abstract: As multimodal language models (MLMs) are increasingly used in social and collaborative settings, it is crucial to evaluate their perspective-taking abilities. Existing benchmarks largely rely on text-based vignettes or static scene understanding, leaving visuospatial perspective-taking...
Compression Method Matters: Benchmark-Dependent Output Dynamics in LLM Prompt Compression
arXiv:2603.23527v1 Announce Type: new Abstract: Prompt compression is often evaluated by input-token reduction, but its real deployment impact depends on how compression changes output length and total inference cost. We present a controlled replication and extension study of benchmark-dependent output...
DepthCharge: A Domain-Agnostic Framework for Measuring Depth-Dependent Knowledge in Large Language Models
arXiv:2603.23514v1 Announce Type: new Abstract: Large Language Models appear competent when answering general questions but often fail when pushed into domain-specific details. No existing methodology provides an out-of-the-box solution for measuring how deeply LLMs can sustain accurate responses under adaptive...
The Diminishing Returns of Early-Exit Decoding in Modern LLMs
arXiv:2603.23701v1 Announce Type: new Abstract: In Large Language Model (LLM) inference, early-exit refers to stopping computation at an intermediate layer once the prediction is sufficiently confident, thereby reducing latency and cost. However, recent LLMs adopt improved pretraining recipes and architectures...
PoliticsBench: Benchmarking Political Values in Large Language Models with Multi-Turn Roleplay
arXiv:2603.23841v1 Announce Type: new Abstract: While Large Language Models (LLMs) are increasingly used as primary sources of information, their potential for political bias may impact their objectivity. Existing benchmarks of LLM social bias primarily evaluate gender and racial stereotypes. When...
Implicit Turn-Wise Policy Optimization for Proactive User-LLM Interaction
arXiv:2603.23550v1 Announce Type: new Abstract: Multi-turn human-AI collaboration is fundamental to deploying interactive services such as adaptive tutoring, conversational recommendation, and professional consultation. However, optimizing these interactions via reinforcement learning is hindered by the sparsity of verifiable intermediate rewards and...
PoiCGAN: A Targeted Poisoning Based on Feature-Label Joint Perturbation in Federated Learning
arXiv:2603.23574v1 Announce Type: new Abstract: Federated Learning (FL), as a popular distributed learning paradigm, has shown outstanding performance in improving computational efficiency and protecting data privacy, and is widely applied in industrial image classification. However, due to its distributed nature,...
BXRL: Behavior-Explainable Reinforcement Learning
arXiv:2603.23738v1 Announce Type: new Abstract: A major challenge of Reinforcement Learning is that agents often learn undesired behaviors that seem to defy the reward structure they were given. Explainable Reinforcement Learning (XRL) methods can answer queries such as "explain this...
Deep Neural Regression Collapse
arXiv:2603.23805v1 Announce Type: new Abstract: Neural Collapse is a phenomenon that helps identify sparse and low rank structures in deep classifiers. Recent work has extended the definition of neural collapse to regression problems, albeit only measuring the phenomenon at the...
Why the Maximum Second Derivative of Activations Matters for Adversarial Robustness
arXiv:2603.23860v1 Announce Type: new Abstract: This work investigates the critical role of activation function curvature -- quantified by the maximum second derivative $\max|\sigma''|$ -- in adversarial robustness. Using the Recursive Curvature-Tunable Activation Family (RCT-AF), which enables precise control over curvature...
Can VLMs Reason Robustly? A Neuro-Symbolic Investigation
arXiv:2603.23867v1 Announce Type: new Abstract: Vision-Language Models (VLMs) have been applied to a wide range of reasoning tasks, yet it remains unclear whether they can reason robustly under distribution shifts. In this paper, we study covariate shifts in which the...
Transcending Classical Neural Network Boundaries: A Quantum-Classical Synergistic Paradigm for Seismic Data Processing
arXiv:2603.23984v1 Announce Type: new Abstract: In recent years, a number of neural-network (NN) methods have exhibited good performance in seismic data processing, such as denoising, interpolation, and frequency-band extension. However, these methods rely on stacked perceptrons and standard activation functions,...
Can we generate portable representations for clinical time series data using LLMs?
arXiv:2603.23987v1 Announce Type: new Abstract: Deploying clinical ML is slow and brittle: models that work at one hospital often degrade under distribution shifts at the next. In this work, we study a simple question -- can large language models (LLMs)...
Stochastic Dimension-Free Zeroth-Order Estimator for High-Dimensional and High-Order PINNs
arXiv:2603.24002v1 Announce Type: new Abstract: Physics-Informed Neural Networks (PINNs) for high-dimensional and high-order partial differential equations (PDEs) are primarily constrained by the $\mathcal{O}(d^k)$ spatial derivative complexity and the $\mathcal{O}(P)$ memory overhead of backpropagation (BP). While randomized spatial estimators successfully reduce...
Chain-of-Authorization: Internalizing Authorization into Large Language Models via Reasoning Trajectories
arXiv:2603.22869v1 Announce Type: new Abstract: Large Language Models (LLMs) have become core cognitive components in modern artificial intelligence (AI) systems, combining internal knowledge with external context to perform complex tasks. However, LLMs typically treat all accessible data indiscriminately, lacking inherent...
Describe-Then-Act: Proactive Agent Steering via Distilled Language-Action World Models
arXiv:2603.23149v1 Announce Type: new Abstract: Deploying safety-critical agents requires anticipating the consequences of actions before they are executed. While world models offer a paradigm for this proactive foresight, current approaches relying on visual simulation incur prohibitive latencies, often exceeding several...
Functional Component Ablation Reveals Specialization Patterns in Hybrid Language Model Architectures
arXiv:2603.22473v1 Announce Type: new Abstract: Hybrid language models combining attention with state space models (SSMs) or linear attention offer improved efficiency, but whether both components are genuinely utilized remains unclear. We present a functional component ablation framework applied to two...
AI Mental Models: Learned Intuition and Deliberation in a Bounded Neural Architecture
arXiv:2603.22561v1 Announce Type: new Abstract: This paper asks whether a bounded neural architecture can exhibit a meaningful division of labor between intuition and deliberation on a classic 64-item syllogistic reasoning benchmark. More broadly, the benchmark is relevant to ongoing debates...
Computational Arbitrage in AI Model Markets
arXiv:2603.22404v1 Announce Type: new Abstract: Consider a market of competing model providers selling query access to models with varying costs and capabilities. Customers submit problem instances and are willing to pay up to a budget for a verifiable solution. An...
Lie to Me: How Faithful Is Chain-of-Thought Reasoning in Reasoning Models?
arXiv:2603.22582v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning has been proposed as a transparency mechanism for large language models in safety-critical deployments, yet its effectiveness depends on faithfulness (whether models accurately verbalize the factors that actually influence their outputs), a...
Beyond Preset Identities: How Agents Form Stances and Boundaries in Generative Societies
arXiv:2603.23406v1 Announce Type: new Abstract: While large language models simulate social behaviors, their capacity for stable stance formation and identity negotiation during complex interventions remains unclear. To overcome the limitations of static evaluations, this paper proposes a novel mixed-methods framework...
Sparse but Critical: A Token-Level Analysis of Distributional Shifts in RLVR Fine-Tuning of LLMs
arXiv:2603.22446v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has significantly improved reasoning in large language models (LLMs), yet the token-level mechanisms underlying these improvements remain unclear. We present a systematic empirical study of RLVR's distributional effects organized...
LGSE: Lexically Grounded Subword Embedding Initialization for Low-Resource Language Adaptation
arXiv:2603.22629v1 Announce Type: new Abstract: Adapting pretrained language models to low-resource, morphologically rich languages remains a significant challenge. Existing vocabulary expansion methods typically rely on arbitrarily segmented subword units, resulting in fragmented lexical representations and loss of critical morphological information....
DALDALL: Data Augmentation for Lexical and Semantic Diverse in Legal Domain by leveraging LLM-Persona
arXiv:2603.22765v1 Announce Type: new Abstract: Data scarcity remains a persistent challenge in low-resource domains. While existing data augmentation methods leverage the generative capabilities of large language models (LLMs) to produce large volumes of synthetic data, these approaches often prioritize quantity...
Analysing LLM Persona Generation and Fairness Interpretation in Polarised Geopolitical Contexts
arXiv:2603.22837v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly utilised for social simulation and persona generation, necessitating an understanding of how they represent geopolitical identities. In this paper, we analyse personas generated for Palestinian and Israeli identities by...