Paired Multimodal Scaling Laws
arXiv:2609.36263v1 Announce Type: cross Abstract: Existing multimodal scaling laws fit multimodality terms empirically after testing and never vary how much data is multimodally paired at fixed data b
Knowledge catalogue
arXiv:2609.36263v1 Announce Type: cross Abstract: Existing multimodal scaling laws fit multimodality terms empirically after testing and never vary how much data is multimodally paired at fixed data b
arXiv:2609.37323v1 Announce Type: new Abstract: Discrete diffusion models have emerged as a powerful paradigm for solving combinatorial optimization (CO) problems on graphs by learning to sample high-
arXiv:2609.37921v1 Announce Type: new Abstract: What are the inductive biases of a Transformer architecture? Existing theory on how the forward pass shapes representations either considers whether Tra
arXiv:2609.37350v1 Announce Type: new Abstract: The development of AI systems for tumor-specific applications is limited by the scarcity of labeled data. Synthetic tumor inpainting offers a promising
arXiv:2609.38023v1 Announce Type: new Abstract: Physics-Informed Neural Networks (PINNs) embed governing equations into deep learning, but enforce them only through loss residuals, leaving highly osci
arXiv:2601.01207v2 Announce Type: replace Abstract: In stochastic message passing, an edge's sampled role changes the node states used to compute adaptive weights at later layers. Weight averaging the
arXiv:2609.36193v1 Announce Type: new Abstract: Learning from scientific measurements often requires aligning modalities with different spatial support and resolution. Subsurface characterization is a
arXiv:2609.36326v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) hands the user's query to whoever hosts the corpus. We propose PILLAR, a Privacy-Preserving RAG (PPRAG) system base
arXiv:2511.22435v2 Announce Type: replace Abstract: Invariant learning on graphs aims to build predictors that rely on causal substructures rather than on environment-specific shortcuts. Current metho
arXiv:2609.37784v1 Announce Type: new Abstract: Satellite observations, precomputed embeddings, and map products describe the same evolving Earth, yet are stored as independent, petabyte-scale data pr
arXiv:2609.36543v1 Announce Type: new Abstract: Partial observability remains a fundamental challenge in robotic navigation, where limited sensor coverage and occlusions leave large portions of the en
arXiv:2609.38054v1 Announce Type: cross Abstract: We present Pow3R-SLAM, a real-time RGB-D simultaneous localization and mapping (SLAM) system that uses Pow3R for tracking and mapping. Inspired by MAS
arXiv:2609.36216v1 Announce Type: new Abstract: Neural operators are typically trained in a supervised fashion, which requires a dataset to be generated with a classical solver. Training them physics-
arXiv:2609.37717v1 Announce Type: cross Abstract: Decoder-only transformers are trained only through a terminal next-token prediction loss, yet this loss constrains every intermediate hidden state thr
arXiv:2609.37789v1 Announce Type: cross Abstract: Self-supervised learning (SSL) by predicting in latent space, without generating the input data itself, learns highly abstract, useful representations
arXiv:2609.37529v1 Announce Type: new Abstract: Pretrained denoisers provide a powerful way to incorporate image priors into restoration algorithms. Plug-and-Play and RED approaches exploit fixed-nois
arXiv:2609.35937v1 Announce Type: cross Abstract: While prior work has documented privacy failures in LLM agents, it remains unclear how the presentation of privacy guidance influences their choice of
arXiv:2609.37202v1 Announce Type: cross Abstract: Droplet collision governs droplet population dynamics in many chemical engineering processes, such as spray drying, spray cooling, agricultural sprayi
arXiv:2609.36279v1 Announce Type: cross Abstract: The shallow embedding of higher-order modal logic in classical higher-order logic, used in Benzmuller and Scott's Notes on Godel's and Scott's variant
arXiv:2603.15106v2 Announce Type: cross Abstract: Enabling efficient deep neural network (DNN) inference on edge devices with different hardware constraints is a challenging task that typically requir
arXiv:2609.36315v1 Announce Type: cross Abstract: Wildfires are an increasing hazard to ecosystems, air quality, and human systems, creating a growing need for datasets that support systematic develop
arXiv:2609.36376v1 Announce Type: cross Abstract: Dense retrieval, the key component of Retrieval Augmented Generation (RAG), retrieves the most relevant documents by comparing dense vector representa
arXiv:2609.36479v1 Announce Type: cross Abstract: Quantum computing has already been explored in several network-security applications. However, how quantum computing may contribute to network-securit
arXiv:2609.36702v1 Announce Type: cross Abstract: Quantum Generative Adversarial Networks (QGANs) have emerged as representative generative models in the Noisy Intermediate-Scale Quantum (NISQ) era an
arXiv:2609.37650v1 Announce Type: new Abstract: Counterfactual explanations of graph neural networks identify edge deletions that flip a prediction. On heterogeneous graphs, however, existing methods
arXiv:2609.36902v1 Announce Type: new Abstract: Existing multimodal fake news detection methods often introduce external information to assist detection. However, most of them rely on entity-level ret
arXiv:2603.17815v2 Announce Type: replace Abstract: Understanding and evaluating multi-step reasoning in LLMs at the level of individual steps remains a key challenge. Process reward models (PRMs) pro
arXiv:2609.37688v1 Announce Type: new Abstract: Reading behaviour varies not only with linguistic input, but also with reader proficiency. In this study, we investigate whether the layer-wise relation
arXiv:2609.36984v1 Announce Type: new Abstract: Complex questions often require multi-hop reasoning that connects facts distributed across sources or distant regions of a long context through intermed
arXiv:2609.36126v1 Announce Type: cross Abstract: Modern AI architectures used to solve visual reasoning tasks typically rely heavily on global connectivity and synchronization. As biological systems
arXiv:2609.38112v1 Announce Type: cross Abstract: Many applications of black-box predictive models require controlling task-relevant error rates, such as missed lesion pixels in segmentation or missed
arXiv:2603.11210v2 Announce Type: replace Abstract: Machine unlearning aims to remove the influence of specific training data from a model while preserving its general utility. In vision, many approxi
arXiv:2609.36452v1 Announce Type: cross Abstract: Masked diffusion language models (MDLMs) can generate text efficiently by predicting multiple masked tokens in parallel, but predictions from the same
arXiv:2609.37587v1 Announce Type: cross Abstract: Longitudinal electronic health record (EHR) modeling requires integrating new visits with an expanding patient history. Yet the continual accumulation
arXiv:2609.35867v1 Announce Type: new Abstract: We propose ReLOBGen, a method for generating limit order book (LOB) messages that are replayable by construction. Replayability is required for closed-l
arXiv:2609.36121v1 Announce Type: cross Abstract: Large language models are vulnerable to prompt injection attacks, where third-party adversarial content can hijack the model's behavior. In this paper
arXiv:2609.36974v1 Announce Type: new Abstract: Text-to-speech models loop, truncate and lose count on text that repeats a phrase many times. We show that repetition itself is what breaks them, not th
arXiv:2609.36916v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) incur high inference latency from long visual token sequences. Existing pruning methods commonly use attention
arXiv:2609.36288v1 Announce Type: new Abstract: EEG foundation models (EEG-FMs) are intended to generalize across different datasets by learning representations that, ideally, are invariant to dataset
arXiv:2609.35890v1 Announce Type: new Abstract: Sparse-autoencoder decomposability and concentrated feature attribution are increasingly treated as evidence that a model's computation is easier to rev
arXiv:2609.36680v1 Announce Type: new Abstract: Visual reprogramming adapts pretrained models to downstream tasks by modifying their input and output interfaces while keeping the backbone fixed. In vi
arXiv:2609.37225v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) have shown strong potential for universal multimodal representation learning. However, existing methods eithe
arXiv:2609.37391v1 Announce Type: new Abstract: Diffusion language models (DLMs) enable parallel generation by predicting and committing multiple tokens at each denoising step, yet they can generate i
arXiv:2609.37879v1 Announce Type: new Abstract: How many tokens from its context does a language model actually use, and what determines that number? We study this question through self-attention. Wit
arXiv:2609.37493v1 Announce Type: cross Abstract: Serving an answer from a large language model requires deciding when to abstain, yet a verifier's ranking accuracy alone does not determine the error
arXiv:2609.37925v1 Announce Type: cross Abstract: Autoregressive (AR) video diffusion enables low-latency, streamable video generation, but prediction errors often accumulate over long rollouts. Train
arXiv:2609.37402v1 Announce Type: new Abstract: Large language model (LLM) routing reduces serving cost by assigning each query to an appropriate model while preserving response quality. Learning such
arXiv:2609.37976v1 Announce Type: cross Abstract: LLMs trained with Chain-of-thought excel in reasoning capability, but often come with excessive token cost. We find that the core of reasoning capacit
arXiv:2609.36031v1 Announce Type: new Abstract: Learning from human videos offers a promising route to acquiring diverse manipulation skills. Extending this capability beyond tabletop settings to long
arXiv:2609.36784v1 Announce Type: new Abstract: When transferring manipulability across systems with different sizes and kinematic structures, matching absolute ellipsoid scale may be unnecessary when
arXiv:2609.37416v1 Announce Type: new Abstract: Post-training quantization (PTQ) methods in the GPTQ family minimize a layer-wise reconstruction error on a uniform grid whose scale must be chosen; the
arXiv:2609.37951v1 Announce Type: new Abstract: Replacing a visible advertisement in a broadcast frame is geometrically straightforward but photometrically delicate. A pasted graphic can have the corr
arXiv:2609.36653v1 Announce Type: new Abstract: Recurrent reasoning models have attracted growing attention for scaling test-time computation, typically by iteratively refining latent states with shar
arXiv:2609.36527v2 Announce Type: new Abstract: Recovering complete physical fields from sparse observations is challenging because the measurements may not uniquely determine the underlying state. Di
arXiv:2609.37917v1 Announce Type: new Abstract: Projection pursuit searches for a direction along which the data look least Gaussian. When the observation space contains a large Gaussian complement, t
arXiv:2609.36600v1 Announce Type: cross Abstract: Classical stochastic approximation methods rely on estimators of the first moment (mean) of a random regression function. We study methods that employ
arXiv:2609.36590v1 Announce Type: new Abstract: Self-speculative decoding accelerates large language model (LLM) inference by drafting tokens from the target model itself, but faces a sharp tradeoff b
arXiv:2411.02562v2 Announce Type: replace Abstract: Segmentation of cellular structures in electron microscopy (EM) images is fundamental to analyzing the morphology of neurons and glial cells in the
arXiv:2609.37040v1 Announce Type: cross Abstract: Natural language autoencoders translate a language model's internal activations into readable explanations. Explaining every token position is costly.
arXiv:2609.37782v1 Announce Type: new Abstract: Large language models (LLMs) have shown strong potential as training-free text encoders for long-context embeddings. Existing approaches primarily impro