daVinci-LLM:Towards the Science of Pretraining

• URL: http://arxiv.org/abs/2603.27164v1
• Date: Sat, 28 Mar 2026 07:00:49 GMT
• Title: daVinci-LLM:Towards the Science of Pretraining
• Abstract summary: daVinci-LLM combines industrial-scale resources with full research freedom to advance the science of pretraining.<n>We release complete data processing pipelines, full training processes, and systematic exploration results.
• Score: 58.6458452580545
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The foundational pretraining phase determines a model's capability ceiling, as post-training struggles to overcome capability foundations established during pretraining, yet it remains critically under-explored. This stems from a structural paradox: organizations with computational resources operate under commercial pressures that inhibit transparent disclosure, while academic institutions possess research freedom but lack pretraining-scale computational resources. daVinci-LLM occupies this unexplored intersection, combining industrial-scale resources with full research freedom to advance the science of pretraining. We adopt a fully-open paradigm that treats openness as scientific methodology, releasing complete data processing pipelines, full training processes, and systematic exploration results. Recognizing that the field lacks systematic methodology for data processing, we employ the Data Darwinism framework, a principled L0-L9 taxonomy from filtering to synthesis. We train a 3B-parameter model from random initialization across 8T tokens using a two-stage adaptive curriculum that progressively shifts from foundational capabilities to reasoning-intensive enhancement. Through 200+ controlled ablations, we establish that: processing depth systematically enhances capabilities, establishing it as a critical dimension alongside volume scaling; different domains exhibit distinct saturation dynamics, necessitating adaptive strategies from proportion adjustments to format shifts; compositional balance enables targeted intensification while preventing performance collapse; how evaluation protocol choices shape our understanding of pretraining progress. By releasing the complete exploration process, we enable the community to build upon our findings and systematic methodologies to form accumulative scientific knowledge in pretraining.

Related papers

• Can In-Context Learning Support Intrinsic Curiosity? [25.43664318918942]
  We investigate whether the emergent in-context learning capabilities of sequence models can eliminate this bottleneck.<n>Specifically, we evaluate whether an exploration policy can be trained to maximize learning progress.<n>We prove a positive result for a broad subclass of non-temporal settings.
  arXiv  Detail & Related papers  (2026-06-17T18:11:06Z)
• A Systematic Post-Train Framework for Video Generation [76.26555417456773]
  Large-scale video diffusion models have demonstrated impressive capabilities in generating high-resolution and semantically rich content.<n>A significant gap remains between their pretraining performance and real-world deployment requirements due to critical issues such as prompt sensitivity, temporal inconsistency, and prohibitive inference costs.<n>We propose a comprehensive post-training framework that systematically aligns pretrained models with user intentions through four synergistic stages.
  arXiv  Detail & Related papers  (2026-04-28T09:34:51Z)
• Evolution of Optimization Methods: Algorithms, Scenarios, and Evaluations [98.44542103979735]
  First-order gradient descent methods, epitomized by gradient descent (SGD) and Adam, serve as the cornerstone of modern training pipelines.<n>Large-scale model training, stringent privacy requirements, and distributed learning paradigms expose critical limitations in these conventional approaches regarding privacy protection and memory efficiency.<n>We retrospectively analyze the evolutionary trajectory of deep learning optimization algorithms and present a comprehensive empirical evaluation of mainstreams across diverse model architectures and training scenarios.<n>We distill key emerging trends and fundamental design trade-offs, pinpointing promising directions for future research.
  arXiv  Detail & Related papers  (2026-04-14T17:01:36Z)
• On the Value of Tokeniser Pretraining in Physics Foundation Models [7.911638286054963]
  We investigate the impact of tokeniser pretraining on the accuracy and efficiency of physics emulation simulations.<n>We show that pretraining the tokeniser with an autoencoding objective prior to training the dynamics model enhances computational efficiency for physics emulation.
  arXiv  Detail & Related papers  (2026-03-05T19:00:22Z)
• LLM-Inspired Pretrain-Then-Finetune for Small-Data, Large-Scale Optimization [7.8639568562295965]
  We consider small-data, large-scale decision problems in which a firm must make many operational decisions simultaneously.<n>We propose a pretrain-then-finetune approach built on a designed Transformer model to address this challenge.
  arXiv  Detail & Related papers  (2026-02-03T16:08:33Z)
• How to Set the Learning Rate for Large-Scale Pre-training? [73.03133634525635]
  We formalize this investigation into two distinct research paradigms: Fitting and Transfer.<n>Within the Fitting Paradigm, we introduce a Scaling Law for search factor, effectively reducing the search complexity from O(n3) to O(n*C_D*C_) via predictive modeling.<n>We extend the principles of $$Transfer to the Mixture of Experts (MoE) architecture, broadening its applicability to encompass model depth, weight decay, and token horizons.
  arXiv  Detail & Related papers  (2026-01-08T15:55:13Z)
• Sample-Efficient Neurosymbolic Deep Reinforcement Learning [49.60927398960061]
  We propose a neuro-symbolic Deep RL approach that integrates background symbolic knowledge to improve sample efficiency.<n>Online reasoning is performed to guide the training process through two mechanisms.<n>We show improved performance over a state-of-the-art reward machine baseline.
  arXiv  Detail & Related papers  (2026-01-06T09:28:53Z)
• From Physics to Machine Learning and Back: Part II - Learning and Observational Bias in PHM [52.64097278841485]
  Review examines how incorporating learning and observational biases through physics-informed modeling and data strategies can guide models toward physically consistent and reliable predictions.<n>Fast adaptation methods including meta-learning and few-shot learning are reviewed alongside domain generalization techniques.
  arXiv  Detail & Related papers  (2025-09-25T14:15:43Z)
• Neuro-Symbolic Predictive Process Monitoring [7.295472411867107]
  We propose a Neuro-Symbolic Predictive Process Monitoring (PPM) approach that integrates data-driven learning with temporal logic-based prior knowledge.<n>We introduce a novel method to incorporate Linear Temporal Logic over finite traces (LTLf) into the training process of autoregressive sequence predictors.<n> Experimental evaluation on three real-world datasets shows that our method improves suffix prediction accuracy and compliance with temporal constraints.
  arXiv  Detail & Related papers  (2025-08-31T13:07:17Z)
• Will Pre-Training Ever End? A First Step Toward Next-Generation Foundation MLLMs via Self-Improving Systematic Cognition [89.50068130832635]
  Self-Improving cognition (SIcog) is a self-learning framework for constructing next-generation foundation MLLMs by multimodal knowledge.<n>We propose Chain-of-Description for step-by-step visual understanding and integrate structured Chain-of-Thought (CoT) reasoning to support in-depth multimodal reasoning.<n>Experiments demonstrate SIcog's effectiveness in developing MLLMs with enhanced multimodal cognition.
  arXiv  Detail & Related papers  (2025-03-16T00:25:13Z)
• Streamflow Prediction with Uncertainty Quantification for Water Management: A Constrained Reasoning and Learning Approach [27.984958596544278]
  This paper studies a constrained reasoning and learning (CRL) approach where physical laws represented as logical constraints are integrated as a layer in the deep neural network. To address small data setting, we develop a theoretically-grounded training approach to improve the generalization accuracy of deep models.
  arXiv  Detail & Related papers  (2024-05-31T18:53:53Z)
• Rethinking Resource Management in Edge Learning: A Joint Pre-training and Fine-tuning Design Paradigm [87.47506806135746]
  In some applications, edge learning is experiencing a shift in focusing from conventional learning from scratch to new two-stage learning. This paper considers the problem of joint communication and computation resource management in a two-stage edge learning system. It is shown that the proposed joint resource management over the pre-training and fine-tuning stages well balances the system performance trade-off.
  arXiv  Detail & Related papers  (2024-04-01T00:21:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.