Search: “pretraining”

arXiv:2610.05292v1 Announce Type: new Abstract: What happens to information a pretrained model already encodes when post-training no longer rewards using it? The common language of representation comp

arXiv:2610.04538v1 Announce Type: new Abstract: Reward-guided image editing at test time seeks to improve a specified reward while preserving source content and visual plausibility. Many existing appr

arXiv:2610.04827v1 Announce Type: new Abstract: Time-series foundation models (TSFMs) enable zero-shot forecasting through large-scale cross-domain pretraining, while retrieval augmentation further im

arXiv:2610.04920v1 Announce Type: new Abstract: Pretrained world models can generate diverse environments, yet users often want to explore a particular scene specified by their own video. This require

arXiv:2610.05329v1 Announce Type: new Abstract: Large language models (LLMs) are typically pretrained in high precision but increasingly deployed with low-precision post-training quantization (PTQ). R

arXiv:2610.02799v1 Announce Type: new Abstract: Fisheye cameras give mobile robots a single-sensor, low-cost view of their surroundings, yet the COCO-pretrained detectors that practitioners routinely

arXiv:2610.01692v1 Announce Type: new Abstract: Wrist electrodermal activity (EDA) differs in amplitude from one person to the next, so affect-recognition models normalize their input before classific

arXiv:2610.01933v1 Announce Type: cross Abstract: Inference-time scaling adapts pretrained diffusion models to new sampling tasks without additional training. Existing methods rely primarily on Monte

arXiv:2610.01712v1 Announce Type: new Abstract: In-context learning (ICL) enables a pretrained model to infer a task from demonstrations without updating its parameters. While much of the existing the

arXiv:2610.00411v1 Announce Type: new Abstract: Contemporary Genomic Foundation Models (GFMs) rely on a DNA-as-a-string paradigm that employs masked token prediction objectives for pretraining. Howeve

arXiv:2609.40333v1 Announce Type: new Abstract: Self-supervised learning draws inspiration from infant visual development, yet standard training pipelines bear little resemblance to it: images are ind

arXiv:2609.39673v1 Announce Type: new Abstract: Can a pretrained graph model replace training and tuning a separate predictor for each dataset? Answering this requires evaluating prediction quality al

arXiv:2609.39420v1 Announce Type: new Abstract: Large language models are strong general-purpose code generators, but executable algorithmic trading remains a demanding specialization target: a model

arXiv:2609.33424v1 Announce Type: cross Abstract: Self-supervised pretraining learns representations that are broadly transferable across downstream tasks, yet direct fine-tuning can be suboptimal due

arXiv:2609.32743v1 Announce Type: new Abstract: Electroencephalography (EEG) foundation models (FMs) promise transferable neural representations, yet their advantages over strong supervised baselines

arXiv:2609.32453v1 Announce Type: new Abstract: Robotic manipulation is inherently history-dependent, yet most pretrained robotic policies condition on only the current observation or a short temporal

arXiv:2609.34371v1 Announce Type: new Abstract: On-policy distillation (OPD) specializes pretrained video diffusion models through teacher supervision along the student's own generation trajectory. Al

arXiv:2609.35473v1 Announce Type: new Abstract: In self-supervised pretraining for Handwritten Text Recognition (HTR), pixel reconstruction methods outperform contrastive methods, unlike in natural-im

arXiv:2609.32756v1 Announce Type: new Abstract: Foundation models are rarely used as generic, frozen feature extractors; instead, they are fine-tuned for the target downstream application. This practi

arXiv:2609.35249v1 Announce Type: cross Abstract: Pretrained robot manipulation policies such as vision-language-action models (VLAs) or world-action models (WAMs) leave interaction-relevant metric ge

arXiv:2609.30173v1 Announce Type: cross Abstract: We introduce GridSFM, a framework that combines a pretrained foundation model across grid topologies with physics-informed fine-tuning for solving AC

arXiv:2609.29405v1 Announce Type: cross Abstract: We adapt Reinforce Adjoint Matching (RAM), a reward-based post-training method, to generative speech enhancement (SE). Starting from a pretrained SE m

arXiv:2607.14236v2 Announce Type: replace-cross Abstract: Pretrained vision-language-action (VLA) policies provide strong language-conditioned manipulation knowledge, but they remain largely vision-dr

arXiv:2609.25298v1 Announce Type: new Abstract: Cultural evaluation coverage and robustness in language models are difficult to diagnose because pretraining corpora and cultural benchmarks are rarely

arXiv:2602.11139v2 Announce Type: replace Abstract: Tabular foundation models, such as TabPFNv2 and TabICL, have recently dethroned gradient-boosted trees at the top of predictive benchmarks, demonstr

arXiv:2609.14829v1 Announce Type: cross Abstract: We introduce Enemray, a Hassaniya-centric language model that enables general-purpose interaction in Hassaniya. Enemray is trained around a stability-

arXiv:2609.14500v1 Announce Type: new Abstract: AI scaling studies increasingly evaluate systems that combine a pretrained model with retrieval, search, verification, tools, and interaction. Yet a hig

arXiv:2609.11897v1 Announce Type: new Abstract: Causal discovery aims to uncover causal structures from data and is fundamental to scientific reasoning and intervention-based decision making. Its eval

arXiv:2609.05662v1 Announce Type: new Abstract: Full-page end-to-end Optical Music Recognition seeks to transcribe entire music pages directly into symbolic notation, avoiding the limitations of tradi

arXiv:2609.05075v1 Announce Type: new Abstract: General continual learning (GCL) aims to learn from evolving data streams without task identities, explicit boundaries, or repeated access to previous d

arXiv:2609.04348v1 Announce Type: new Abstract: Object-centric visual representations are important for physical-world perception, but existing visual pretraining methods often capture semantic catego

arXiv:2602.01734v2 Announce Type: replace Abstract: Training instability remains a critical challenge in large language model (LLM) pretraining, often manifesting as sudden gradient explosions that wa

arXiv:2609.01657v1 Announce Type: cross Abstract: Multimodal models often build on architectures designed for generative vision-language modeling, typically combining separately pretrained vision enco

arXiv:2609.02531v1 Announce Type: new Abstract: World Action Models (WAMs) leverage the capabilities of large-scale pretrained video diffusion models to jointly predict future observations and actions

arXiv:2609.02546v1 Announce Type: new Abstract: Zero-shot generalization to unseen embodiments is important for generalizable vision-language-action (VLA) models as robot hardware evolves and task-spe

arXiv:2609.00518v1 Announce Type: new Abstract: Antibody-specific language models pretrained via masked language modeling (MLM) learn representations that are critical for downstream sequence design a

arXiv:2609.01341v1 Announce Type: cross Abstract: Do multilingual language models transfer factual knowledge across languages during continued pretraining, or do they mostly recall facts learned direc

arXiv:2608.30092v1 Announce Type: cross Abstract: We present Arkios, a 1.04B-parameter dense transformer pretrained from scratch on 150B tokens of bilingual English-Nepali text, using a custom single-

arXiv:2608.26288v1 Announce Type: new Abstract: Muon has emerged as a strong optimizer for the matrix-valued parameters in large language model pretraining, approximately orthogonalizing its momentum

arXiv:2608.25677v1 Announce Type: new Abstract: Parameter-efficient fine-tuning is often assumed to preserve pretrained capabilities because it updates only a small number of parameters. We show that

arXiv:2608.25693v1 Announce Type: new Abstract: Acquiring pixel-level annotations for medical image segmentation is a severe bottleneck. Traditional U-Net architectures, while effective, learn local t

arXiv:2608.24364v1 Announce Type: new Abstract: Self-supervised pretraining enables transferable representations for medical imaging, yet most CT encoders remain biased toward coarse semantic understa

COMPASS is a scalable cross‑embodiment robot navigation framework that builds on NVIDIA’s X‑Mobility pretrained policies, adding lightweight RL fine‑tuning specialists for new robot‑scene pairs to avo

arXiv:2608.21571v1 Announce Type: new Abstract: Lung cancer remains a leading cause of cancer-related mortality worldwide, and early diagnosis is critical for improving survival. However, early-stage

arXiv:2608.19540v1 Announce Type: cross Abstract: Training fast generators on new domains with limited data remains challenging for two reasons. First, adapting a pretrained diffusion or flow model to

arXiv:2606.20151v2 Announce Type: replace-cross Abstract: This work proposes a hybrid ANN-SNN pipeline that effectively leverages the rich embeddings of pretrained artificial neural networks (ANNs) to