9734 papers
This paper proposes "Interactive Benchmarks," a unified evaluation paradigm that assesses model intelligence through active information acquisition and reasoning under budget constraints in interactive proofs and games, demonstrating that current models still have significant room for improvement in these dynamic scenarios.
This paper introduces CONE, a hybrid transformer encoder that utilizes a novel composite embedding algorithm to preserve the semantics of units and variables for complex numerical data, achieving state-of-the-art performance in numerical reasoning tasks across diverse domains.
KindSleep is a deep learning framework that integrates clinical knowledge with single-channel oximetry and patient data to accurately diagnose obstructive sleep apnea, achieving superior performance and enhanced transparency across large, diverse datasets.
This landscape commentary evaluates the GPT-5 family against GPT-4o, revealing substantial improvements in expert-level textual reasoning and multimodal synthesis that approach state-of-the-art performance in tasks like mammography, while highlighting that generalist models still lag behind specialized systems in perception-critical domains such as neuroradiology.
This paper introduces ConTSG-Bench, a unified benchmark featuring a large-scale, multi-modal dataset and comprehensive metrics to systematically evaluate and analyze the performance, limitations, and future directions of conditional time series generation models.
This paper proposes a distributional risk-sensitive reinforcement learning framework that integrates Information Bottleneck representations and Conditional Value-at-Risk optimization to achieve certified worst-case DRAM equalizer performance with significant speedups and uncertainty quantification, outperforming existing methods by up to 89.1% on real-world memory data.
This paper presents the first structural-assumption-free causal discovery method for linear non-Gaussian latent-variable cyclic models by establishing a graphical criterion for distributional equivalence, introducing edge rank constraints, and providing an algorithm to recover models up to this equivalence class.
This paper introduces a novel and efficient on-policy reinforcement learning method that trains diffusion policies by aligning policy iteration with the diffusion process, thereby overcoming computational bottlenecks in log-likelihood estimation while enabling multimodal behavior generation and entropy regularization across diverse benchmark tasks.
This paper proposes Diffusion Contrastive Reconstruction (DCR), a method that injects contrastive signals derived from reconstructed images into the diffusion process to resolve gradient conflicts and jointly optimize both discriminative and detail-perceptive abilities, thereby overcoming the limitations of CLIP's visual encoder for balanced visual representation.
This paper demonstrates that the architectural inductive biases of locality and weight sharing in convolutional neural networks fundamentally alter implicit regularization by coupling learned filters to low-dimensional patch manifolds, thereby enabling generalization on high-dimensional spherical data where fully connected networks provably fail.
This paper presents WhisperAlign, a solution for the DL Sprint 4.0 that combines word-boundary-aware ASR using whisper-timestamped chunking and domain-fine-tuned Pyannote diarization anchored by WhisperX to achieve high-accuracy transcription and speaker separation for long-form Bengali speech.
This paper establishes a unified framework linking quadratic polarities to deformed Legendre transformations via linear algebra on homogeneous coordinates, defines polar divergences that generalize Fenchel-Young and Bregman divergences, and elucidates the reference duality in information geometry through total polar Fenchel-Young divergences.
This paper investigates the generalization capabilities of a multimodal foundation model fine-tuned on diverse synthetic interactive data for the novel task of Open-Set Corrective Assistance, demonstrating that effective open-set assistive intelligence requires datasets encompassing multimodal grounding, defect inference, and exposure to varied scenarios.
This paper proposes the Class-specific Augmentation based Disentanglement (CAD) framework to mitigate instance entanglement in instance-dependent partial label learning by employing intra-class feature alignment and inter-class weighted penalty mechanisms to clarify class boundaries and reduce confusion.
This paper introduces a multilevel training framework for Kolmogorov-Arnold Networks (KANs) that leverages their structural equivalence to multichannel MLPs and the properties of spline basis functions to create a properly nested hierarchy of models, resulting in orders-of-magnitude improvements in training accuracy and speed, particularly for physics-informed neural networks.
This paper introduces MCal, a lightweight post-hoc method that effectively corrects missingness bias in feature attribution explanations by fine-tuning a simple linear head on frozen models, outperforming or matching expensive retraining approaches across diverse medical benchmarks.
This paper proves that standard RLHF alignment is inherently shallow because gradient signals vanish once a sequence's harmfulness is determined, and it proposes a recovery penalty objective to ensure alignment gradients persist throughout the entire generation process.
This paper introduces Osmosis Distillation, a novel model hijacking attack that exploits synthetic datasets generated by dataset distillation methods to compromise deep learning models in transfer learning with high success rates using only a few poisoned samples while maintaining utility on original tasks.
This paper introduces ReCouPLe, a lightweight framework that leverages natural language rationales as causal guidance to train reward models that are robust to spurious correlations and capable of zero-shot transfer to novel tasks, significantly outperforming baselines in reward accuracy and downstream policy performance under distribution shifts.
This paper presents a scalable, interpretable framework that achieves 80.4% accuracy in classifying eight professional baseball pitch types using only monocular 3D body kinematics, revealing that upper-body mechanics—particularly wrist position and trunk tilt—are the primary predictors while establishing an empirical ceiling for grip-based distinctions.