deep-neural-networks

Documents about deep-neural-networks

• Towards accurate extreme event likelihoods from diffusion model climate emulators#155298arXiv CS
• Training-Free Probabilistic Time-Series Forecasting with Conformal Seasonal Pools#155300arXiv CS
• UniPool: A Globally Shared Expert Pool for Mixture-of-Experts#168298arXiv CS
• Directional Consistency as a Complementary Optimization Signal: The GONO Framework#168330arXiv CS
• Diverse Sampling in Diffusion Models with Marginal Preserving Particle Guidance#168337arXiv CS
• Sequential Design of Genetic Circuits Under Uncertainty With Reinforcement Learning#168338arXiv CS
• Hedging Memory Horizons for Non-Stationary Prediction via Online Aggregation#168339arXiv CS
• LoKA: Low-precision Kernel Applications for Recommendation Models At Scale#175250arXiv CS
• Masked Generative Transformer Is What You Need for Image Editing#175255arXiv CS
• MaD Physics: Evaluating information seeking under constraints in physical environments#175264arXiv CS
• ConQuR: Corner Aligned Activation Quantization via Optimized Rotations for LLMs#175273arXiv CS
• DynaMiCS: Fine-tuning LLMs with Performance Constraints using Dynamic Mixtures#175280arXiv CS
• Dynamic Cross-Modal Prompt Generation for Multimodal Continual Instruction Tuning#175281arXiv CS
• A Semi-Supervised Framework for Speech Confidence Detection using Whisper#178886arXiv CS
• MetaColloc: Optimization-Free PDE Solving via Meta-Learned Basis Functions#178892arXiv CS
• A New Technique for AI Explainability using Feature Association Map#178894arXiv CS
• Multi-Variable Conformal Prediction: Optimizing Prediction Sets without Data Splitting#178896arXiv CS
• Transferable Delay-Aware Reinforcement Learning via Implicit Causal Graph Modeling#178903arXiv CS
• In-context learning to predict critical transitions in dynamical systems#178904arXiv CS
• What is Learnable in Valiant's Theory of the Learnable?#180637arXiv CS
• Image Based Helmet Detection Using Deep Neural Networks#126655Zenodo (CERN)
• Meta-learning In-Context Enables Training-Free Cross Subject Brain Decoding#2581arXiv CS
• Demystifying OPD: Length Inflation and Stabilization Strategies for Large Language Models#2582arXiv CS
• What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal#2583arXiv CS
• TTVS: Boosting Self-Exploring Reinforcement Learning via Test-time Variational Synthesis#2589arXiv CS
• Adversarial Label Invariant Graph Data Augmentations for Out-of-Distribution Generalization#2597arXiv CS
• SOLAR: Communication-Efficient Model Adaptation via Subspace-Oriented Latent Adapter Reparametrization#2600arXiv CS
• Scaling-Aware Data Selection for End-to-End Autonomous Driving Systems#2601arXiv CS
• Equivariant Efficient Joint Discrete and Continuous MeanFlow for Molecular Graph Generation#2615arXiv CS
• A Direct Approach for Handling Contextual Bandits with Latent State Dynamics#2619arXiv CS
• Small Vision-Language Models are Smart Compressors for Long Video Understanding#2621arXiv CS
• Initialisation Determines the Basin: Efficient Codebook Optimisation for Extreme LLM Quantization#2622arXiv CS
• Large Language Models Generate Harmful Content Using a Distinct, Unified Mechanism#5943arXiv CS
• Case-Grounded Evidence Verification: A Framework for Constructing Evidence-Sensitive Supervision#5945arXiv CS
• Biologically-Grounded Multi-Encoder Architectures as Developability Oracles for Antibody Design#5967arXiv CS
• Distributed Online Convex Optimization with Compressed Communication: Optimal Regret and Applications#5978arXiv CS
• ClawGUI: A Unified Framework for Training, Evaluating, and Deploying GUI Agents#10333arXiv CS
• RationalRewards: Reasoning Rewards Scale Visual Generation Both Training and Test Time#10346arXiv CS
• CAGenMol: Condition-Aware Diffusion Language Model for Goal-Directed Molecular Generation#10368arXiv CS
• From Attribution to Action: A Human-Centered Application of Activation Steering#10371arXiv CS
• OSC: Hardware Efficient W4A4 Quantization via Outlier Separation in Channel Dimension#13095arXiv CS
• Information-Theoretic Optimization for Task-Adapted Compressed Sensing Magnetic Resonance Imaging#13102arXiv CS
• BID-LoRA: A Parameter-Efficient Framework for Continual Learning and Unlearning#13103arXiv CS
• From Imitation to Discrimination: Progressive Curriculum Learning for Robust Web Navigation#13104arXiv CS
• TimeSAF: Towards LLM-Guided Semantic Asynchronous Fusion for Time Series Forecasting#13108arXiv CS
• SOAR: Self-Correction for Optimal Alignment and Refinement in Diffusion Models#13111arXiv CS
• Orthogonal Subspace Projection for Continual Machine Unlearning via SVD-Based LoRA#13114arXiv CS
• Complex Interpolation of Matrices with an application to Multi-Manifold Learning#14005arXiv CS
• Hierarchical Reinforcement Learning with Runtime Safety Shielding for Power Grid Operation#14015arXiv CS
• Provably Efficient Offline-to-Online Value Adaptation with General Function Approximation#14026arXiv CS