4/13/2026
Do LLMs Follow Their Own Rules? A Reflexive Audit of Self-Stated Safety Policies
Symbolic-Neural Consistency Audit (SNCA) framework that extracts LLM self-stated safety policies via prompts and verifies model adherence to them.
4/13/2026
Natural Riemannian gradient for learning functional tensor networks
Riemannian gradient descent approach for optimizing low-rank functional tensor networks on arbitrary loss functions beyond least-squares regression.
4/13/2026
Online Intention Prediction via Control-Informed Learning
Online intention prediction framework for autonomous systems using inverse reinforcement learning with time-varying objectives and unknown parameters.
4/13/2026
Iterative Identification Closure: Amplifying Causal Identifiability in Linear SEMs
Iterative Identification Closure framework for determining causal identifiability in linear structural equation models with latent confounders.
4/13/2026
Transferable FB-GNN-MBE Framework for Potential Energy Surfaces: Data-Adaptive Transfer Learning in Deep Learned Many-Body Expansion Theory
Fragment-based graph neural network integrated with many-body expansion theory for predicting potential energy surfaces in chemical systems.
4/13/2026
Biologically-Grounded Multi-Encoder Architectures as Developability Oracles for Antibody Design
CrossAbSense framework using protein language model encoders and attention decoders to predict antibody properties for therapeutic design validation.
4/13/2026
Variational Quantum Physics-Informed Neural Networks for Hydrological PDE-Constrained Learning with Inherent Uncertainty Quantification
Hybrid quantum-classical physics-informed neural networks for hydrological modeling with uncertainty quantification using variational quantum circuits.
4/13/2026
Sharp description of local minima in the loss landscape of high-dimensional two-layer ReLU neural networks
Theoretical analysis of loss landscape in two-layer ReLU neural networks, characterizing local minima and their connection to stochastic gradient descent dynamics.
4/13/2026
Beyond Augmented-Action Surrogates for Multi-Expert Learning-to-Defer
Learning-to-Defer framework that routes inputs to experts while selecting additional information (retrieved documents, tool outputs) to provide each expert, extending traditional routing systems.
4/13/2026
Automated Instruction Revision (AIR): A Structured Comparison of Task Adaptation Strategies for LLM
Systematic comparison of LLM task adaptation strategies including instruction revision, prompt optimization, and retrieval methods.
4/13/2026
RecaLLM: Addressing the Lost-in-Thought Phenomenon with Explicit In-Context Retrieval
Post-training method enabling LLMs to retrieve and reason over long-context information effectively.
4/13/2026
Envisioning the Future, One Step at a Time
Video prediction model representing scene dynamics as sparse point trajectories for efficient future frame synthesis.
4/13/2026
Case-Grounded Evidence Verification: A Framework for Constructing Evidence-Sensitive Supervision
Framework for training LLMs to make evidence-dependent predictions by grounding supervision in case-specific evidence.
4/13/2026
Large Language Models Generate Harmful Content Using a Distinct, Unified Mechanism
Mechanistic study using weight pruning to identify unified internal mechanism LLMs use for generating harmful content.
4/13/2026
Low Rank Based Subspace Inference for the Laplace Approximation of Bayesian Neural Networks
Method using low-rank techniques for Bayesian uncertainty quantification in neural networks via Laplace approximation.
4/13/2026
Neurons Speak in Ranges: Breaking Free from Discrete Neuronal Attribution
Research on polysemanticity in LLMs showing neurons encode multiple concepts, challenging discrete attribution methods for model interpretability.
4/13/2026
Task-agnostic Low-rank Residual Adaptation for Efficient Federated Continual Fine-Tuning
Federated continual fine-tuning with low-rank residual adaptation, enabling efficient parameter-efficient learning across new classes in federated settings.
4/13/2026
Revitalizing Black-Box Interpretability: Actionable Interpretability for LLMs via Proxy Models
Proxy model framework for efficient post-hoc interpretability of LLMs, reducing computational costs of model-agnostic explanations.
4/13/2026
Provable Post-Training Quantization: Theoretical Analysis of OPTQ and Qronos
Theoretical analysis of OPTQ/GPTQ post-training quantization for LLMs, providing rigorous quantitative guarantees for PTQ algorithms.
4/13/2026
On-the-Fly Adaptation to Quantization: Configuration-Aware LoRA for Efficient Fine-Tuning of Quantized LLMs
Configuration-aware LoRA adaptation for quantized LLMs enabling efficient edge device deployment with heterogeneous capabilities.
4/13/2026
Large Reasoning Models Learn Better Alignment from Flawed Thinking
RECAP: RL method for safety alignment in large reasoning models, teaching critical evaluation of flawed premises via counter-aligned prefilling.
4/13/2026
LLM4Delay: Flight Delay Prediction via Cross-Modality Adaptation of Large Language Models and Aircraft Trajectory Representation
LLM-based flight delay prediction integrating textual aeronautical data and aircraft trajectories for air traffic management.
4/13/2026
Dual Mamba for Node-Specific Representation Learning: Tackling Over-Smoothing with Selective State Space Modeling
Graph neural network architecture using selective state space modeling to address over-smoothing in deep GNNs via node-specific representation evolution.
4/13/2026
Boosting Brain-inspired Path Integration Efficiency via Learning-based Replication of Continuous Attractor Neurodynamics
Optimization of continuous attractor neural networks for brain-inspired path integration, reducing computational redundancy in navigation systems.
4/13/2026
AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention
Vision-Language-Action model with active visual attention for robotic manipulation, extending from Markov to partially observable decision processes.
4/13/2026
Adaptive Tuning of Parameterized Traffic Controllers via Multi-Agent Reinforcement Learning
Multi-agent RL framework for adaptive traffic signal control, replacing static controllers with learning-based optimization for complex traffic dynamics.
4/13/2026
Bandwidth-constrained Variational Message Encoding for Cooperative Multi-agent Reinforcement Learning
Multi-agent RL for graph-based coordination with bandwidth constraints, addressing what information agents should transmit under communication limits.
4/13/2026
The Two-Stage Decision-Sampling Hypothesis: Understanding the Emergence of Self-Reflection in RL-Trained LLMs
Analysis of self-reflection emergence in LLMs through RL post-training, using gradient attribution to explain distinct solution generation and revision capabilities.
4/13/2026
Imitation Learning for Combinatorial Optimisation under Uncertainty
Imitation learning framework for combinatorial optimization problems, examining how expert demonstrations affect policy learning in sequential decision problems.
4/13/2026
FP8-RL: A Practical and Stable Low-Precision Stack for LLM Reinforcement Learning
FP8 low-precision quantization for LLM reinforcement learning, addressing memory and compute bottlenecks in rollout generation with engineering and algorithmic solutions.
4/13/2026
On the Limits of Layer Pruning for Generative Reasoning in Large Language Models
Demonstrates layer pruning limitations for LLM reasoning tasks, showing pruned models lose algorithmic capabilities despite compression on classification tasks.
4/13/2026
dnaHNet: A Scalable and Hierarchical Foundation Model for Genomic Sequence Learning
dnaHNet foundation model for genomic sequence learning with tokenizer-free design preserving biological motifs while handling long contexts efficiently.
4/13/2026
Reinforcement-aware Knowledge Distillation for LLM Reasoning
Reinforcement-aware knowledge distillation method for distilling RL-trained reasoning LLMs into smaller models while preserving chain-of-thought capability.
4/13/2026
Accelerating Local LLMs on Resource-Constrained Edge Devices via Distributed Prompt Caching
Distributed prompt caching technique for accelerating local LLM inference on resource-constrained edge devices via inter-device state sharing.
4/13/2026
Implicit Bias in Deep Linear Discriminant Analysis
Analyzes implicit regularization of Deep LDA objective for scale-invariant discriminative metric learning.
4/13/2026
Why Adam Can Beat SGD: Second-Moment Normalization Yields Sharper Tails
Theoretical analysis explaining Adam's empirical advantage over SGD through second-moment normalization using stopping-time/martingale analysis.
4/13/2026
Training event-based neural networks with exact gradients via Differentiable ODE Solving in JAX
Enables exact gradient computation for spiking neural networks via differentiable ODE solving in JAX, supporting arbitrary neuron models.
4/13/2026
Memory-efficient Continual Learning with Prototypical Exemplar Condensation
Proposes prototypical exemplar condensation for memory-efficient continual learning, reducing stored samples per class from 20+ to single digits.
4/13/2026
ALMAB-DC: Active Learning, Multi-Armed Bandits, and Distributed Computing for Sequential Experimental Design and Black-Box Optimization
ALMAB-DC framework combines active learning, multi-armed bandits, and distributed computing for expensive black-box optimization.
4/13/2026
Mechanisms of Introspective Awareness
Investigates mechanisms of introspective awareness in LLMs, where models detect injected steering vectors with minimal false positives.
4/13/2026
Boosted Distributional Reinforcement Learning: Analysis and Healthcare Applications
Analyzes distributional reinforcement learning with applications to healthcare, moving beyond expectation-based objectives for uncertain domains.
4/13/2026
Hierarchical SVG Tokenization: Learning Compact Visual Programs for Scalable Vector Graphics Modeling
Proposes hierarchical SVG tokenization approach for improved scalable vector graphics modeling with LLMs via geometric-aware token design.
4/13/2026
ALTO: Adaptive LoRA Tuning and Orchestration for Heterogeneous LoRA Training Workloads
ALTO system for adaptive hyperparameter tuning and orchestration of LoRA fine-tuning jobs across heterogeneous multi-tenant environments.
4/13/2026
Conformal Margin Risk Minimization: An Envelope Framework for Robust Learning under Label Noise
Proposes CMRM, a framework for improving classification under label noise without privileged knowledge, using quantile-calibrated regularization.
4/13/2026
BLEG: LLM Functions as Powerful fMRI Graph-Enhancer for Brain Network Analysis
Combines LLMs with Graph Neural Networks to enhance fMRI brain network analysis by leveraging LLM representations.
4/13/2026
Constraining Sequential Model Editing with Editing Anchor Compression
Method for constraining sequential editing of LLMs to prevent knowledge degradation using editing anchor compression.
4/13/2026
Gen-n-Val: Agentic Image Data Generation and Validation
Agentic system for generating and validating synthetic image data to address data scarcity and label noise in vision tasks.
4/13/2026
Bayesian Social Deduction with Graph-Informed Language Models
Evaluates LLM reasoning capabilities in social deduction game Avalon using Bayesian inference with graph-informed models.
4/13/2026
Bayesian Ego-graph Inference for Networked Multi-Agent Reinforcement Learning
arXiv paper on Bayesian ego-graph inference for decentralized multi-agent reinforcement learning with constrained communication.
4/13/2026
Interactive Program Synthesis for Modeling Collaborative Physical Activities from Narrated Demonstrations
arXiv paper on interactive program synthesis for collaborative physical task modeling from narrated demonstrations.