4540 papers
This paper proposes a curriculum-based continual learning framework that decomposes complex robust control problems with multiple uncertainties into sequential tasks, combining a model-based controller with deep reinforcement learning to achieve efficient, non-forgetting policy updates and successful sim-to-real transfer for automotive powertrain vibration control.
The paper introduces Coupled Discrete Diffusion (CoDD), a hybrid framework that overcomes the "factorization barrier" in diffusion language models by replacing fully factorized outputs with a lightweight probabilistic inference layer, thereby enabling efficient parallel generation of coherent, high-quality text without the prohibitive costs of full joint modeling or reinforcement learning.
This paper introduces SpeedTransformer, a novel Transformer-based model that utilizes only speed inputs from dense smartphone GPS trajectories to achieve superior accuracy and transferability in transportation mode detection compared to traditional deep learning approaches.
This paper establishes that history-dependent policies with sublinear expected regret are robust-optimal for non-rectangular average-reward robust MDPs without requiring rectangularity, and introduces a transient-value framework with an epoch-based policy that achieves constant-order finite-time performance by combining worst-case optimality with online learning.
The paper introduces DUEL, a framework that enables exact likelihood computation for masked diffusion models under the test-time distribution, revealing that their true performance significantly surpasses previous estimates and establishing a new standard for comparing and optimizing parallel text generation.
This paper introduces \textsc{Gome}, a gradient-based MLE agent that outperforms traditional tree search methods on MLE-Bench by mapping diagnostic reasoning to gradient computation, demonstrating that as LLM reasoning capabilities improve, gradient-based optimization becomes increasingly superior to exhaustive enumeration.
The paper introduces FinTexTS, a large-scale financial text-paired time-series dataset constructed via a novel semantic-based and multi-level pairing framework that overcomes the limitations of simple keyword matching by leveraging LLMs to align news articles with stock prices across macro, sector, related company, and target-company levels, thereby significantly improving stock price forecasting performance.
This paper introduces two software-only techniques, Overflow-Aware Scaling (OAS) and Macro Block Scaling (MBS), that significantly reduce the accuracy gap between the hardware-efficient MXFP4 format and NVIDIA's NVFP4 standard in Large Language Models, achieving near-parity performance with minimal computational overhead.
This paper proposes the Online-Within-Online Fair Multi-Task Learning (OWO-FMTL) framework, which leverages a dual-loop mechanism with primal-dual updates to ensure long-term equitable inference performance for heterogeneous users in AI-RANs while maintaining low computational overhead.
KernelCraft introduces the first benchmark evaluating agentic LLM systems that use feedback-driven workflows to automatically generate and optimize low-level kernels for emerging hardware with novel ISAs, demonstrating their ability to produce valid, high-performance code that rivals or exceeds traditional compiler baselines.
This paper presents ALADIN, an accuracy-latency-aware framework that enables the pre-deployment evaluation of mixed-precision quantized neural networks on scratchpad-based embedded AI accelerators by transforming models into platform-aware representations to analyze trade-offs and bottlenecks without requiring physical hardware.
This paper reviews the landscape of ultra-low-power edge and in-sensor AI processors and empirically benchmarks a segmentation model on GAP9, STM32N6, and Sony IMX500 platforms to demonstrate that while in-sensor processing offers superior energy-delay performance, different architectures provide distinct trade-offs between latency, energy efficiency, and power budgets.
This paper presents a data-rate-aware CNN accelerator architecture for FPGAs that utilizes multi-pixel processing and design-space exploration to optimize hardware utilization and resource efficiency across varying data rates, thereby enabling the efficient implementation of complex CNNs on a single device.
This paper demonstrates that synergistically integrating Supervised Contrastive Learning, Hopfield networks, and Hierarchical Gated Recurrent Networks into Spiking Neural Networks achieves optimal neuromorphic vision performance on N-MNIST by balancing accuracy, energy efficiency, and structured neuronal clustering, rather than relying on isolated architectural optimizations.
This paper introduces Hebbian-Oscillatory Co-Learning (HOC-L), a unified two-timescale framework that integrates hyperbolic sparse geometry with Kuramoto-based phase synchronization to enable connectivity consolidation only when phase coherence signals meaningful patterns, thereby achieving provable convergence and complexity in bio-inspired neural architectures.
This paper introduces Auralink SDC, an edge-deployed multi-agent AI architecture that autonomously manages electric vehicle charging infrastructure with high reliability and sub-50ms latency, achieving 78% autonomous incident resolution and 87.6% diagnostic accuracy to address the critical failure rates and slow remediation times of current cloud-centric systems.
This paper presents a sensitivity-guided framework for compressing Reservoir Computing accelerators that systematically balances quantization and pruning to significantly improve hardware efficiency and reduce power consumption on FPGAs while maintaining high model accuracy across various time-series tasks.
The AetherFloat Family introduces a novel block-scale-free, quad-radix floating-point architecture that eliminates the hardware overhead of dynamic scaling and IEEE 754 inefficiencies, achieving significant area, power, and latency reductions in AI accelerators through explicit mantissas, base-4 scaling, and stochastic rounding.
This paper presents a physics-informed neural network (PINN) framework that robustly reconstructs hidden state variables and estimates biophysical parameters in multiscale neuronal models using only partial, noisy voltage observations, effectively overcoming the convergence failures and sensitivity issues common in traditional numerical methods.
This paper introduces the Variable-Invariant Two-Dimensional State Space Model (VI 2D SSM) and its unified VI 2D Mamba architecture, which theoretically establish and implement a permutation-equivariant framework for multivariate time series that eliminates artificial variable ordering to achieve state-of-the-art performance with improved structural scalability.