3797 papers
This paper introduces Code-Space Response Oracles (CSRO), a novel framework that replaces black-box deep reinforcement learning oracles with Large Language Models to generate human-readable, interpretable multi-agent policies as code, achieving competitive performance while enabling the discovery of complex, explainable strategies.
This paper introduces BlueDown, a new post-processing algorithm that leverages succinct block hierarchical regression to produce more accurate and consistent demographic estimates for the US Census while maintaining the same privacy guarantees and structural constraints as the existing TopDown method.
This paper proposes a hardware-efficient "soft sparsity" paradigm for CNNs that utilizes a Most Significant Bit (MSB) proxy to skip negligible non-zero multiplications, achieving significant MAC and power reductions with zero accuracy loss while outperforming traditional zero-skipping methods.
The paper introduces CLIPO, a method that integrates contrastive learning into policy optimization to generalize Reinforcement Learning with Verifiable Rewards (RLVR) by capturing invariant structures across correct reasoning paths, thereby mitigating hallucinations and improving the generalization and robustness of Large Language Models.
This paper argues that the "Lost in the Middle" phenomenon in large language models is an inherent geometric property of causal decoder architectures present at initialization, caused by the interplay of causal masking and residual connections that creates a structurally hostile "dead zone" in the middle of the context, a bias that persists even after standard pretraining.
This paper proposes an unsupervised hyperspectral image clustering method that improves upon existing Wasserstein space dictionary learning by utilizing unbalanced Wasserstein barycenters to learn a robust lower-dimensional representation, thereby mitigating issues of class blurring and sensitivity to outliers and noise.
This paper proposes a neural operator integrated with an implicit numerical scheme that learns underlying state-space dynamics from limited data to accurately predict vibration frequency response curves for engineered components without relying on physics-based regularizing loss functions.
The paper proposes Mashup Learning, a method that accelerates LLM finetuning and improves downstream accuracy by identifying and merging relevant historical checkpoints to serve as an optimized initialization for new tasks, thereby reducing training time by up to 37% compared to training from scratch.
This paper proposes ReMix, a novel Mixture-of-LoRAs framework that employs non-learnable routing weights and a Reinforce Leave-One-Out (RLOO) gradient estimator to prevent routing imbalance, thereby ensuring all active LoRAs contribute equally and significantly outperforming state-of-the-art parameter-efficient finetuning methods.
This paper introduces DT-BEHRT, a graph-enhanced transformer model that improves predictive performance and interpretability in electronic health record analysis by explicitly modeling disease trajectories within organ systems and employing a novel pre-training strategy to capture heterogeneous medical code interactions.
This paper establishes a general stability criterion for stochastic mirror descent algorithms to enable valid statistical inference in adaptive bandit settings, introducing regularized-EXP3 variants that simultaneously achieve minimax-optimal regret, nominal confidence interval coverage, and robustness to adversarial corruptions.
This paper introduces ARCHE, an end-to-end learned image compression framework that achieves state-of-the-art rate-distortion efficiency by unifying hierarchical, spatial, and channel-based priors with adaptive feature recalibration, all while maintaining computational efficiency without relying on recurrent or transformer-based components.
This paper introduces Adaptive Activation Cancellation (AAC), a real-time, training-free inference framework that mitigates hallucinations in large language models by identifying and suppressing hallucination-associated neural activations as structured interference, thereby improving factual accuracy across multiple model scales without degrading general capabilities or fluency.
This paper proposes actor-accelerated Policy Dual Averaging, a method that employs a learned policy network to efficiently approximate optimization sub-problems in continuous action spaces, thereby maintaining theoretical convergence guarantees while achieving superior performance over standard on-policy baselines like PPO.
This paper proposes a hybrid Hidden Markov Model that combines Laplace quantile-defined market states with a Poisson-driven jump-duration mechanism to generate synthetic equity excess growth rates that simultaneously preserve heavy-tailed distributions, volatility clustering, and realistic tail-state dwell times, outperforming standard GARCH and HMM models in joint distributional and temporal fidelity.
This paper introduces Flexible Cutoff Learning (FCL), a method that trains machine learning interatomic potentials with randomly sampled cutoff radii to enable post-training optimization of per-atom cutoffs, thereby significantly reducing computational costs for specific applications without requiring retraining.
The paper introduces FusionNet, a neural network that reconstructs high-temporal-resolution 4D cardiac motion from short-duration CMR scans by estimating intermediate 3D heart shapes, achieving superior accuracy with a Dice coefficient exceeding 0.897 compared to existing methods.
The paper introduces SDSR, a scalable spectral divide-and-conquer algorithm for species tree reconstruction that achieves up to 10-fold faster runtimes compared to standard methods while maintaining comparable accuracy under the multispecies coalescent model.
This paper establishes that a continuous-time diffusion approximation of policy gradient for stochastic bandits achieves logarithmic regret with a learning rate of , while demonstrating that a significantly smaller learning rate of is necessary to avoid linear regret in specific instances.
This paper extends the harmonic loss framework by systematically evaluating various non-Euclidean distance metrics across vision and language models, demonstrating that cosine-based variants offer superior trade-offs in accuracy, interpretability, and sustainability compared to traditional cross-entropy and Euclidean approaches.