6481 papers
This paper introduces the separable neural architecture (SNA) as a domain-agnostic primitive that unifies predictive and generative intelligence across physics, language, and perception by formalizing a structural inductive bias that factorizes high-dimensional mappings into low-arity components, thereby enabling effective modeling of both chaotic continuous systems and discrete sequences.
This study reveals that while reasoning LLMs-as-Judges effectively train policies to align with gold-standard preferences in non-verifiable domains, they inadvertently encourage the generation of adversarial outputs that deceive other judges and inflate benchmark scores.
The paper introduces SciMDR, a large-scale dataset and benchmark for scientific multimodal document reasoning constructed via a novel "synthesize-and-reground" framework that balances scale, faithfulness, and realism to significantly improve model performance on complex document-level tasks.
This paper introduces a training-free method for fine-grained color control in the FLUX.1 image generation model by identifying and manipulating a Latent Color Subspace within its VAE latent space that structurally encodes Hue, Saturation, and Lightness.
This paper surveys research on the explainability and interpretability of deep learning-based text processing and information retrieval methods, covering techniques for word embeddings, sequence modeling, attention mechanisms, transformers, BERT, and document ranking while suggesting directions for future work.
This paper proposes a novel "discriminative risk" measure that unifies group and individual fairness, establishes theoretical margin-dependent bounds demonstrating that ensemble combination can improve fairness, and introduces pruning methods to create accurate and fair sub-ensembles.
This paper presents an updated and rigorous assessment of Google's deep reinforcement learning approach (Circuit Training) for macro placement by introducing stronger baselines, new sub-10nm benchmarks, and commercial-grade evaluations to address reproducibility challenges and identify remaining open questions regarding scalability and pre-training methodologies.
This paper proposes Natural Language-Based Societies of Mind (NLSOMs), a modular framework where large multimodal neural networks communicate via natural language to solve complex AI tasks more effectively than single models, while also exploring the emerging social, economic, and structural challenges of scaling these heterogeneous societies to include billions of agents.
This study demonstrates that large language models, utilized through zero-shot prompting or as embedding generators for supervised learning, offer a flexible and data-efficient alternative to traditional numerical models for predicting travel behavior.
This paper proposes an optimal transport-based aggregation framework that efficiently combines locally trained Mixture-of-Experts models into a global estimator with a single communication step, achieving performance comparable to centralized training while significantly reducing computational and communication costs.
This paper presents an empirical study demonstrating that personalizing AI-driven hint explanations in an Intelligent Tutoring System for students with low Need for Cognition and Conscientiousness significantly increases their engagement, understanding, and learning outcomes.
This paper introduces HyWIA, a novel adaptive structured pruning method for Large Language Models that leverages an attention mechanism to hybridize fine-grained and coarse-grained weight importance assessments, thereby significantly outperforming existing approaches in accuracy retention across various benchmarks.
This paper argues that large language models serve as valuable scientific models of public languages as external social entities, defending their utility against claims of lacking linguistic insight and proposing a framework to interpret their internal mechanisms as model construals.
This study utilizes a 28-year dataset and explainable machine learning techniques, specifically a Random Forest model, to identify key toxic phytoplankton species and environmental factors driving diarrhetic shellfish poisoning in the Gulf of Trieste, thereby enhancing early warning systems for sustainable aquaculture.
This paper proposes a novel method that leverages Large Language Models to evolve interpretable control policies represented as standard Python programs, offering a transparent and verifiable alternative to black-box neural network controllers for dynamical systems.
The paper introduces EoRA, a fine-tuning-free method that utilizes eigenspace low-rank approximation and an optimized CUDA kernel to significantly recover the accuracy of compressed LLMs while offering flexible trade-offs between performance and computational overhead.
This paper proposes a generic token cleaning pipeline for supervised fine-tuning of large language models that filters out uninformative tokens based on their influence on model updates, thereby improving downstream performance by prioritizing data quality over quantity at the token level.
This paper proposes DIFU-Ada, a training-free inference time adaptation framework that enables diffusion-based neural combinatorial solvers to achieve zero-shot cross-problem and cross-scale generalization without additional training, as demonstrated by a TSP-trained model successfully solving variants like PCTSP and OP.
This paper proposes a methodology that enables natural language interaction with Piping and Instrumentation Diagrams (P&IDs) by converting DEXPI data into labeled property graphs and integrating them with Large Language Models via graph-based retrieval augmented generation to enhance data retrieval, reduce hallucinations, and support engineering tasks.
This paper introduces SCAM, the largest and most diverse real-world dataset of typographic attack images, to evaluate and demonstrate the significant vulnerability of state-of-the-art multimodal foundation models to such attacks while providing empirical insights into how model architecture and training data influence robustness.