5028 papers
DevBench is a realistic, telemetry-driven benchmark comprising 1,800 instances across six languages that evaluates LLMs on code completion tasks with a focus on ecological validity, contamination-free assessment, and detailed diagnostic insights to guide practical model selection and development.
This paper presents the first component-level survey systematically reviewing the bidirectional interactions between Large Language Models and Multi-Armed Bandits, highlighting how MAB algorithms optimize LLM workflows while LLMs redefine core MAB components to enhance adaptive decision-making.
ELSA is a novel framework that integrates split learning and hierarchical federated learning with client clustering, dynamic model splitting, and privacy-preserving communication sketches to enable efficient, robust, and privacy-aware fine-tuning of large language models on resource-constrained edge networks.
This paper proposes a novel data-rate-aware continuous-flow architecture for CNN inference on FPGAs that mitigates hardware underutilization caused by data reduction in pooling and strided convolution layers by interleaving signals and sharing resources, thereby enabling the high-throughput implementation of complex models like MobileNet on a single device.
MeanCache is a training-free framework that accelerates Flow Matching inference by replacing instantaneous velocity caching with an average-velocity approach using cached Jacobian-vector products and a trajectory-stability scheduling strategy, achieving significant speedups (up to 4.56X) while maintaining high generation quality across models like FLUX.1 and HunyuanVideo.
PASS is a scalable framework for pairwise-constrained k-means clustering that optimizes a small working subset while formally certifying the feasibility of cannot-link constraints via list-coloring, thereby enabling efficient classical and quantum solutions with verifiable repair mechanisms for infeasible instances.
This paper presents a systematic empirical comparison demonstrating that model-free neural estimators, particularly state-space models, achieve state estimation performance comparable to strong nonlinear Kalman filters in nonlinear dynamical systems while offering significantly higher inference throughput.
TimeSliver is a novel explainable deep learning framework for time-series classification that jointly leverages raw data and symbolic abstraction to linearly encode temporal segment contributions, achieving superior attribution accuracy and competitive predictive performance compared to existing methods.
This paper proposes TGCC, a novel causal-invariance-based graph dataset condensation method that extracts domain-invariant features and injects them via spectral contrastive learning to significantly improve performance in cross-task and cross-domain scenarios while maintaining state-of-the-art results in single-task settings.
FlowSymm is a novel graph attention architecture that recovers missing network flows by combining a physics-aware, symmetry-preserving group-action framework with a lightweight Tikhonov refinement, ensuring exact adherence to local conservation laws while outperforming state-of-the-art methods across transportation, energy, and mobility benchmarks.
Mem-T introduces an autonomous memory agent trained via the MoT-GRPO framework, which densifies sparse long-horizon rewards through tree-guided backpropagation to achieve superior performance and efficiency compared to existing memory management systems.
This paper proposes a Bitcoin price prediction model using Combinatorial Fusion Analysis (CFA) to integrate diverse machine learning models via rank-score characteristics and weighted combinations, achieving a superior Mean Absolute Percentage Error (MAPE) of 0.19% that outperforms individual models and existing prediction methods.
This paper introduces Adam-Aware In-Run Data Shapley, a novel method that overcomes the limitations of SGD-based attribution in adaptive optimizers by deriving a closed-form approximation and a Linearized Ghost Approximation to achieve near-perfect fidelity in data contribution estimation while maintaining high training efficiency.
This paper investigates whether Schwartz higher-order values improve sentence-level human value detection, finding that while hierarchical gating offers limited benefits, calibration techniques and hybrid ensembles significantly boost performance, suggesting the value hierarchy is more effective as an inductive bias than a rigid routing mechanism.
This paper introduces LatentMem, a learnable multi-agent memory framework that addresses memory homogenization and information overload by using an experience bank and a memory composer to generate customized, token-efficient latent memories, further optimized via Latent Memory Policy Optimization (LMPO) to significantly enhance multi-agent system performance.
This paper introduces T2T (Thickening-to-Thinning), a dynamic reward shaping framework inspired by human learning dynamics that enhances LLM reasoning by encouraging longer, exploratory trajectories on incorrect attempts and penalizing length upon success, thereby outperforming standard baselines on mathematical benchmarks.
This paper introduces a novel inference-time backdoor attack that exploits maliciously modified chat templates to compromise open-weight language models without altering weights or training data, demonstrating high success rates in degrading factual accuracy and inducing harmful outputs while evading current security scans.
This paper introduces the Hinge Regression Tree (HRT), a novel oblique decision tree method that reframes split learning as a non-linear least-squares problem solvable via a damped Newton method, offering provable convergence, universal approximation capabilities, and superior performance with compact structures compared to existing baselines.
This paper introduces Radial Müntz-Szász Networks (RMN), a highly parameter-efficient neural architecture that utilizes learnable radial power bases and a log-primitive to accurately model multidimensional singular fields like $1/r\log r$, achieving significantly lower error rates than standard MLPs and SIREN on benchmark tasks while providing closed-form gradients for physics-informed learning.
SDFed is a heterogeneous federated prompt learning framework that addresses local-global discrepancies by combining a fixed-length global prompt with variable-length local prompts, enhanced by subspace refinement and divergence control strategies to improve performance and robustness in privacy-sensitive, resource-constrained multi-party settings.