Knowledge catalogue
arXiv:2609.33523v1 Announce Type: new Abstract: We present In-Token Learning, an image restoration framework that adapts a pretrained diffusion transformer using conditional rectified flow matching. C
arXiv:2609.33874v1 Announce Type: new Abstract: JET uses pretrained language and vision-language models to select among a finite set of answers without additional training. It evaluates candidate like
arXiv:2609.35297v1 Announce Type: new Abstract: Pretraining a language model takes enormous compute, and the right optimizer can save a good part of it. Muon's spectral step gives a stronger direction
arXiv:2509.23235v2 Announce Type: replace Abstract: Model inversion is a widely adopted technique in data-free learning that reconstructs synthetic inputs from a pretrained model through iterative opt
arXiv:2609.34292v1 Announce Type: cross Abstract: We report a three-month autonomous research-agent program testing five solid-state-physics-inspired compression mappings on pretrained language models
arXiv:2609.34061v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models integrate pretrained Vision-Language Models (VLMs) with action heads for robot control. Common action heads have d
arXiv:2609.32263v1 Announce Type: new Abstract: Multimodal test-time adaptation (TTA) aims to adapt a pretrained multimodal model online to distribution shift across modalities using unlabeled test da
arXiv:2609.32108v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models leverage large-scale pretraining to ultimately achieve generalist manipulation. Deployed VLA policies must support c
arXiv:2510.15508v2 Announce Type: replace Abstract: In this study, we propose an enhancement to the similarity computation mechanism in multimodal contrastive pretraining frameworks such as CLIP. Prio
arXiv:2609.34642v1 Announce Type: new Abstract: Schrodinger bridges provide an entropy-regularized framework and a principled solution for unpaired domain translation. In practice, a pretrained bridge
arXiv:2609.35649v1 Announce Type: new Abstract: Tandem mass spectrum prediction supports compound identification across metabolomics, natural-product discovery, and environmental analysis. However, pr
arXiv:2609.33359v1 Announce Type: new Abstract: Geometric view synthesis can expand a single wine-label photograph into a training set, but its value with pretrained image encoders is unclear. We stud
arXiv:2505.09738v2 Announce Type: replace-cross Abstract: Pretrained language models (LLMs) are often constrained by their fixed tokenization schemes, leading to inefficiencies and performance limitat
arXiv:2609.31546v1 Announce Type: new Abstract: Electrocardiogram (ECG) foundation models typically tokenize the signal into fixed-length patches that ignore cardiac structure, so a patch may split a
arXiv:2609.31140v1 Announce Type: new Abstract: Most Vision-Language Models (VLMs) are built by extending pretrained Large Language Models (LLMs) with visual modules and multimodal alignment. However,
arXiv:2609.30880v1 Announce Type: new Abstract: Time series foundation models (TSFMs) are pretrained on series from diverse domains, where demand series make up only a small fraction. Demand data has
arXiv:2609.31204v1 Announce Type: cross Abstract: Recent fMRI foundation models differ substantially in the spatial scale at which they represent brain activity. ROI- and connectivity-based models are
arXiv:2609.31558v1 Announce Type: new Abstract: Contrastive Language--Image Pretraining (CLIP) has emerged as a dominant vision backbone due to its strong transferability and zero-shot capabilities. H
arXiv:2609.30957v1 Announce Type: new Abstract: The empirical success of pretraining large language models has inspired a deeper investigation into the underlying loss landscapes and the optimization
arXiv:2609.31448v1 Announce Type: cross Abstract: Clinical prediction models estimate risk from patient measurements, while large language models support medical text understanding and question answer
arXiv:2609.29630v1 Announce Type: cross Abstract: Recent topic models leverage pretrained embeddings, but neural architectures produce latent representations without grounding in specific texts, and c
arXiv:2609.28807v1 Announce Type: new Abstract: Over the course of a lifetime, robots may encounter novel scenarios unaccounted for in its original training that result in performance degradation. One
arXiv:2609.28811v1 Announce Type: new Abstract: World-action models (WAMs) transfer visual and motion priors from pretrained video generators to robot control by jointly modeling visual dynamics and a
arXiv:2609.29027v1 Announce Type: new Abstract: Atmospheric observations are sparse, heterogeneous, and unevenly distributed, whereas many generative atmospheric models learn distributions over regula
arXiv:2609.29505v1 Announce Type: new Abstract: Diffusion inference repeatedly evaluates the same large network. We ask whether pretrained weights alone can identify residual branches that need not be
arXiv:2609.28838v1 Announce Type: cross Abstract: Online reinforcement learning fine-tuning of pretrained flow-matching vision-language-action (VLA) policies promises robots that keep learning after d
arXiv:2605.06192v2 Announce Type: replace-cross Abstract: Pretrained video diffusion models provide powerful spatiotemporal generative priors, making them a natural foundation for robotic world models
arXiv:2609.27264v1 Announce Type: new Abstract: We present GaussPDE, a framework that injects physically structured partial differential equation (PDE) dynamics into pretrained 3D Gaussian scenes with
arXiv:2609.27963v1 Announce Type: new Abstract: Few-step generation with pretrained diffusion and flow models can be accelerated by lightweight training that optimizes the sampling trajectory rather t
arXiv:2609.28473v1 Announce Type: new Abstract: Representation Autoencoders (RAEs) enable diffusion models to operate in the feature spaces of pretrained visual encoders. However, many off-the-shelf e
arXiv:2609.27510v1 Announce Type: cross Abstract: Modern large language models are pretrained on massive datasets, making it difficult to prevent benchmark data from entering their training sets and u
arXiv:2609.25058v1 Announce Type: new Abstract: Parameter-efficient fine-tuning (PEFT) adapts large language models (LLMs) to downstream tasks while updating only a small fraction of their pretrained
arXiv:2609.26605v1 Announce Type: cross Abstract: Foundation model embeddings of screening mammograms may encode pre-diagnostic tissue change without task-specific adaptation. We tested whether embedd
arXiv:2609.25972v1 Announce Type: new Abstract: NAWE (Neural-Assisted Watermark Extraction) combines an explicit signal-processing watermarking construction with a pretrained neural host predictor. A
arXiv:2609.26071v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models inherit strong semantic priors from large-scale vision-language pretraining, yet remain limited in robotic manipulat
arXiv:2609.24565v1 Announce Type: new Abstract: Architectural drawings encode material classes through repeated hatch patterns. We test whether a connectome-constrained fly visual network, pretrained
arXiv:2609.24170v1 Announce Type: new Abstract: Embodied AI systems are often organized into System 1 and System 2. System 1 is typically a pretrained policy that generates actions at high frequency,
arXiv:2609.24525v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have demonstrated remarkable generalization in robotic manipulation via large-scale multimodal pretraining. However,
arXiv:2609.24976v1 Announce Type: cross Abstract: Dexterous manipulation depends on contact dynamics that are often only partially observable from vision. Recent World-Action Models (WAMs) couple pred
arXiv:2609.24868v1 Announce Type: new Abstract: World Action Models (WAMs) jointly generate robot actions and predict future world states, transferring priors from video pretraining to robot control.
arXiv:2609.23983v1 Announce Type: new Abstract: Neuroimaging foundation models pretrained on large, predominantly western cohorts are increasingly proposed as general-purpose backbones for brain MRI a
arXiv:2609.23863v1 Announce Type: new Abstract: Manipulation policies must know which objects matter and where they are, yet the pretrained backbones that current robot foundation models build on, fro
arXiv:2609.24271v1 Announce Type: new Abstract: Current embodied systems largely rely on pretrained capabilities that remain fixed after deployment, limiting their ability to learn from physical inter
arXiv:2609.24517v1 Announce Type: cross Abstract: Model merging aims to combine multiple fine-tuned models derived from a common pretrained model into a single multi-task model without additional join
arXiv:2609.24788v1 Announce Type: new Abstract: In this paper, we propose SVEET, a framework that requires merely training on a pretrained bidirectional video diffusion model but supports high-quality
arXiv:2609.24303v1 Announce Type: cross Abstract: Post-training often improves task performance but can degrade confidence calibration, leaving post-trained language models (PoLMs) more overconfident
arXiv:2609.22005v1 Announce Type: cross Abstract: Gating the value pathway of attention reportedly improves language model pretraining, and prior studies disagree on why. We argue and provide experime
arXiv:2609.21022v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have shown strong generalization in robotic manipulation by combining semantic knowledge from pretrained vision-lang
arXiv:2609.21228v1 Announce Type: cross Abstract: Vision-language-action (VLA) models built on pretrained vision-language models have demonstrated strong performance across diverse robotic manipulatio
arXiv:2609.21932v1 Announce Type: new Abstract: Joint remaining useful life (RUL) prediction and capacity estimation require representations of both gradual degradation and recent battery behavior. Th
arXiv:2609.21126v1 Announce Type: new Abstract: We propose a decoupled, layerwise method for structurally sparsifying the fully connected layers of pretrained neural networks. Rather than penalizing a
arXiv:2609.21316v1 Announce Type: new Abstract: Human-readable maps provide an intuitive interface for specifying robot destinations, but connecting their schematic geometry to egocentric observations
arXiv:2609.21220v1 Announce Type: new Abstract: Generative robot policies can represent diverse, multimodal behaviors, but adapting pretrained policies to deployment-time constraints such as collision
arXiv:2609.21740v1 Announce Type: new Abstract: Latent world models enable planning by predicting the effects of actions in a learned representation space, but their predictions can become unreliable
arXiv:2609.20064v1 Announce Type: new Abstract: Despite impressive reported scores, large vision-language models have seen limited practical uptake in historical automatic text recognition because of
arXiv:2609.20106v1 Announce Type: new Abstract: Robotic reward models evaluate task execution from visual observations, but their predictions can change with camera viewpoint and occlusion even when t
arXiv:2609.20751v1 Announce Type: new Abstract: Adapting a pretrained autoregressive (AR) model is a cost-efficient route to a diffusion language model (DLM). While nearly all such adaptations start f
arXiv:2609.19846v1 Announce Type: cross Abstract: As generalist robot policies gain vision and language from web-scale pretraining, demonstrations remain costly to collect and tied to the robot that r
arXiv:2609.20277v1 Announce Type: new Abstract: World Action Models (WAMs) have demonstrated strong robotic manipulation capabilities by augmenting pretrained video generative models with action exper
arXiv:2602.01346v2 Announce Type: replace Abstract: While open sourced Vision-Language Models (VLMs) have proliferated, selecting the optimal pretrained model for a specific downstream task remains ch