{"total":100,"rows":[{"id":"2506.10943","title":"Self-Adapting Language Models","created":"2025-06-12","cited_by_count":84,"area":null},{"id":"2504.13161","title":"Nemotron-CLIMB: CLustering-based Iterative Data Mixture Bootstrapping for Language Model Pre-training","created":"2025-04-17","cited_by_count":66,"area":null},{"id":"2508.15884","title":"Jet-Nemotron: Efficient Language Model with Post Neural Architecture Search","created":"2025-08-21","cited_by_count":41,"area":null},{"id":"2506.04689","title":"Recycling the Web: A Method to Enhance Pre-training Data Quality and Quantity for Language Models","created":"2025-06-05","cited_by_count":31,"area":null},{"id":"2509.14786","title":"Pre-training under infinite compute","created":"2025-09-18","cited_by_count":29,"area":null},{"id":"2412.15285","title":"Maximize Your Data's Potential: Enhancing LLM Accuracy with Two-Phase Pretraining","created":"2024-12-18","cited_by_count":26,"area":null},{"id":"2510.06826","title":"Mid-Training of Large Language Models: A Survey","created":"2025-10-08","cited_by_count":21,"area":null},{"id":"2502.18934","title":"Kanana: Compute-efficient Bilingual Language Models","created":"2025-02-26","cited_by_count":20,"area":null},{"id":"2504.08719","title":"SWAN-GPT: An Efficient and Scalable Approach for Long-Context Language Modeling","created":"2025-04-11","cited_by_count":19,"area":null},{"id":"2507.12466","title":"Language Models Improve When Pretraining Data Matches Target Tasks","created":"2025-07-16","cited_by_count":19,"area":null},{"id":"2505.23725","title":"MuLoCo: Muon is a practical inner optimizer for DiLoCo","created":"2025-05-29","cited_by_count":18,"area":null},{"id":"2507.03253","title":"RefineX: Learning to Refine Pre-training Data at Scale from Expert-Guided Programs","created":"2025-07-04","cited_by_count":17,"area":null},{"id":"2505.18091","title":"Data Mixing Can Induce Phase Transitions in Knowledge Acquisition","created":"2025-05-23","cited_by_count":16,"area":null},{"id":"2510.06213","title":"Training Dynamics Impact Post-Training Quantization Robustness","created":"2025-10-07","cited_by_count":15,"area":null},{"id":"2511.18903","title":"How Learning Rate Decay Wastes Your Best Data in Curriculum-Based LLM Pretraining","created":"2025-11-24","cited_by_count":14,"area":null},{"id":"2601.02204","title":"NextFlow: Unified Sequential Modeling Activates Multimodal Understanding and Generation","created":"2026-01-05","cited_by_count":14,"area":null},{"id":"2504.08716","title":"ModernBERT or DeBERTaV3? Examining Architecture and Data Influence on Transformer Encoder Models Performance","created":"2025-04-11","cited_by_count":12,"area":null},{"id":"2502.04235","title":"Reformulation for Pretraining Data Augmentation","created":"2025-02-06","cited_by_count":10,"area":null},{"id":"2507.10583","title":"$\\texttt{Droid}$: A Resource Suite for AI-Generated Code Detection","created":"2025-07-11","cited_by_count":9,"area":null},{"id":"2602.05879","title":"EuroLLM-22B: Technical Report","created":"2026-02-05","cited_by_count":9,"area":null},{"id":"2506.14111","title":"Essential-Web v1.0: 24T tokens of organized web data","created":"2025-06-17","cited_by_count":8,"area":null},{"id":"2510.03871","title":"Optimal Scaling Needs Optimal Norm","created":"2025-10-04","cited_by_count":8,"area":null},{"id":"2510.27055","title":"Detecting Data Contamination in LLMs via In-Context Learning","created":"2025-10-30","cited_by_count":8,"area":null},{"id":"2502.14907","title":"GneissWeb: Preparing High Quality Data for LLMs at Scale","created":"2025-02-19","cited_by_count":7,"area":null},{"id":"2511.01066","title":"HPLT 3.0: Very Large-Scale Multilingual Resources for LLMs and MT. Mono- and Bi-lingual Data, Multilingual Evaluation, and Pre-Trained Models","created":"2025-11-02","cited_by_count":7,"area":null},{"id":"2602.07824","title":"Data Darwinism Part I: Unlocking the Value of Scientific Data for Pre-training","created":"2026-02-08","cited_by_count":7,"area":null},{"id":"2603.18534","title":"Data-efficient pre-training by scaling synthetic megadocs","created":"2026-03-19","cited_by_count":7,"area":null},{"id":"2504.01542","title":"Register Always Matters: Analysis of LLM Pretraining Data Through the Lens of Language Variation","created":"2025-04-02","cited_by_count":6,"area":null},{"id":"2507.13390","title":"PARAM-1 BharatGen 2.9B Model","created":"2025-07-16","cited_by_count":6,"area":null},{"id":"2509.09009","title":"Open-sci-ref-0.01: open and reproducible reference baselines for language model and dataset comparison","created":"2025-09-10","cited_by_count":6,"area":null},{"id":"2510.10681","title":"RePro: Training Language Models to Faithfully Recycle the Web for Pretraining","created":"2025-10-12","cited_by_count":6,"area":null},{"id":"2510.24626","title":"Relative Scaling Laws for LLMs","created":"2025-10-28","cited_by_count":6,"area":null},{"id":"2512.06266","title":"Nanbeige4-3B Technical Report: Exploring the Frontier of Small Language Models","created":"2025-12-06","cited_by_count":6,"area":null},{"id":"2602.12414","title":"propella-1: Multi-Property Document Annotation for LLM Data Curation at Scale","created":"2026-02-12","cited_by_count":6,"area":null},{"id":"2603.17837","title":"The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning","created":"2026-03-18","cited_by_count":6,"area":null},{"id":"2604.13977","title":"How Can We Synthesize High-Quality Pretraining Data? A Systematic Study of Prompt Design, Generator Model, and Source Data","created":"2026-04-15","cited_by_count":6,"area":null},{"id":"2605.12715","title":"Scaling Laws for Mixture Pretraining Under Data Constraints","created":"2026-05-12","cited_by_count":6,"area":null},{"id":"2601.09066","title":"Mi:dm 2.0 Korea-centric Bilingual Language Models","created":"2026-01-14","cited_by_count":5,"area":null},{"id":"2506.00469","title":"Massively Multilingual Adaptation of Large Language Models Using Bilingual Translation Data","created":"2025-05-31","cited_by_count":4,"area":null},{"id":"2508.20869","title":"OLMoASR: Open Models and Data for Training Robust Speech Recognition Models","created":"2025-08-28","cited_by_count":4,"area":null},{"id":"2511.04234","title":"Reusing Pre-Training Data at Test Time is a Compute Multiplier","created":"2025-11-06","cited_by_count":4,"area":null},{"id":"2511.07464","title":"Motif 2 12.7B technical report","created":"2025-11-07","cited_by_count":4,"area":null},{"id":"2511.16397","title":"AICC: Parse HTML Finer, Make Models Better -- A 7.3T AI-Ready Corpus Built by a Model-Based HTML Parser","created":"2025-11-20","cited_by_count":4,"area":null},{"id":"2602.09003","title":"Data Science and Technology Towards AGI Part I: Tiered Data Management","created":"2026-02-09","cited_by_count":4,"area":null},{"id":"2602.16687","title":"Scaling Open Discrete Audio Foundation Models with Interleaved Semantic, Acoustic, and Text Tokens","created":"2026-02-18","cited_by_count":4,"area":null},{"id":"2606.16246","title":"Demystifying Training-Time Augmentation for Data-Constrained Language Model Pretraining","created":"2026-06-15","cited_by_count":4,"area":null},{"id":"2609.21967","title":"NemotronLabs VoiceChat: An Open Full-duplex Speech-to-Speech Model with Tool Calling Capabilities","created":"2026-09-18","cited_by_count":4,"area":null},{"id":"2412.15113","title":"Associative memory inspires improvements for in-context learning using a novel attention residual stream architecture","created":"2024-12-19","cited_by_count":3,"area":null},{"id":"2508.01326","title":"Large-Scale Diverse Synthesis for Mid-Training","created":"2025-08-02","cited_by_count":3,"area":null},{"id":"2508.11857","title":"SupraTok: Cross-Boundary Tokenization for Enhanced Language Model Performance","created":"2025-08-16","cited_by_count":3,"area":null},{"id":"2510.20860","title":"Data-Centric Lessons To Improve Speech-Language Pretraining","created":"2025-10-22","cited_by_count":3,"area":null},{"id":"2601.11667","title":"Distill-then-Replace: Efficient Task-Specific Hybrid Attention Model Construction","created":"2026-01-16","cited_by_count":3,"area":null},{"id":"2602.15210","title":"\\\"UberWeb: Insights from Multilingual Curation for a 20-Trillion-Token Dataset","created":"2026-02-16","cited_by_count":3,"area":null},{"id":"2604.08519","title":"Cram Less to Fit More: Training Data Pruning Improves Memorization of Facts","created":"2026-04-09","cited_by_count":3,"area":null},{"id":"2605.10129","title":"Synthetic Pre-Pre-Training Improves Language Model Robustness to Noisy Pre-Training Data","created":"2026-05-11","cited_by_count":3,"area":null},{"id":"2608.17050","title":"Frozen Memory Is Not Enough: Rethinking External Memory as Extraction","created":"2026-08-17","cited_by_count":3,"area":null},{"id":"2508.01317","title":"LinkQA: Synthesizing Diverse QA from Multiple Seeds Strongly Linked by Knowledge Points","created":"2025-08-02","cited_by_count":2,"area":null},{"id":"2512.07612","title":"PCMind-2.1-Kaiyuan-2B Technical Report","created":"2025-12-08","cited_by_count":2,"area":null},{"id":"2604.05688","title":"Attention Editing: A Versatile Framework for Cross-Architecture Attention Conversion","created":"2026-04-07","cited_by_count":2,"area":null},{"id":"2605.17849","title":"Generating Pretraining Tokens from Organic Data for Data-Bound Scaling","created":"2026-05-18","cited_by_count":2,"area":null},{"id":"2605.30202","title":"A Dual-Path Architecture for Scaling Compute and Capacity in LLMs","created":"2026-05-28","cited_by_count":2,"area":null},{"id":"2607.27230","title":"Multi-Head Attention Residuals","created":"2026-07-22","cited_by_count":2,"area":null},{"id":"2510.07227","title":"Where to Begin: Efficient Pretraining via Subnetwork Selection and Distillation","created":"2025-10-08","cited_by_count":1,"area":null},{"id":"2511.18054","title":"Blu-WERP (Web Extraction and Refinement Pipeline): A Scalable Pipeline for Preprocessing Large Language Model Datasets","created":"2025-11-22","cited_by_count":1,"area":null},{"id":"2603.19149","title":"Optimal Splitting of Language Models from Mixtures to Specialized Domains","created":"2026-03-19","cited_by_count":1,"area":null},{"id":"2605.19568","title":"m3BERT: A Modern, Multi-lingual, Matryoshka Bidirectional Encoder","created":"2026-05-19","cited_by_count":1,"area":null},{"id":"2605.23721","title":"Is a Document Educational or Just Wikipedia-Style? -- Pitfalls of Classifier-Based Quality Filtering","created":"2026-05-21","cited_by_count":1,"area":null},{"id":"2607.24717","title":"DataOrchestra: Learning to Orchestrate Per-Example Curation of Pretraining Data","created":"2026-07-27","cited_by_count":1,"area":null},{"id":"2608.09093","title":"The Announcement Carries the Cue: Markup, Boundaries, and the Notation of Pre-Training Corpora","created":"2026-08-10","cited_by_count":1,"area":null},{"id":"2608.28308","title":"Deriving Scaling Laws for OpenEuroLLM Models: Learning Rate, Batch Size and Loss","created":"2026-08-28","cited_by_count":1,"area":null},{"id":"2609.17699","title":"NeMo Data Designer: An Extensible Framework for Multimodal Synthetic Data Generation","created":"2026-09-15","cited_by_count":1,"area":null},{"id":"2509.24356","title":"Beyond Repetition: Text Simplification and Curriculum Learning for Data-Constrained Pretraining","created":"2025-09-29","cited_by_count":0,"area":null},{"id":"2510.01249","title":"LOCA: Logical Chain Augmentation for Scientific Corpus Cleaning","created":"2025-09-24","cited_by_count":0,"area":null},{"id":"2512.07522","title":"LIME: Making LLM Data More Efficient with Linguistic Metadata Embeddings","created":"2025-12-08","cited_by_count":0,"area":null},{"id":"2512.13330","title":"FIN-bench-v2: A Unified and Robust Benchmark Suite for Evaluating Finnish Large Language Models","created":"2025-12-15","cited_by_count":0,"area":null},{"id":"2603.15953","title":"A Family of LLMs Liberated from Static Vocabularies","created":"2026-03-16","cited_by_count":0,"area":null},{"id":"2603.24826","title":"Synthetic Rewriting as a Quality Multiplier: Evidence from Portuguese Continued Pretraining","created":"2026-03-25","cited_by_count":0,"area":null},{"id":"2603.27164","title":"daVinci-LLM:Towards the Science of Pretraining","created":"2026-03-28","cited_by_count":0,"area":null},{"id":"2604.06829","title":"WRAP++: Web discoveRy Amplified Pretraining","created":"2026-04-08","cited_by_count":0,"area":null},{"id":"2605.01077","title":"Teaching LLMs Brazilian Healthcare: Injecting Knowledge from Official Clinical Guidelines","created":"2026-05-01","cited_by_count":0,"area":null},{"id":"2605.10777","title":"Locking Pretrained Weights via Deep Low-Rank Residual Distillation","created":"2026-05-11","cited_by_count":0,"area":null},{"id":"2605.18801","title":"Position: Let's Develop Data Probes to Fundamentally Understand How Data Affects LLM Performance","created":"2026-05-11","cited_by_count":0,"area":null},{"id":"2605.22287","title":"SciCore-Mol: Augmenting Large Language Models with Pluggable Molecular Cognition Modules","created":"2026-05-21","cited_by_count":0,"area":null},{"id":"2606.03825","title":"Dynamic Short Convolutions Improve Transformers","created":"2026-06-02","cited_by_count":0,"area":null},{"id":"2606.22079","title":"Where Does the Signal Live? A Web Data Recipe for Medical Encoder Pretraining","created":"2026-06-20","cited_by_count":0,"area":null},{"id":"2606.30175","title":"CORTEX: High-Quality Cross-Domain Organization of Web-Scale Corpora through Ontological Corpus Graph","created":"2026-06-29","cited_by_count":0,"area":null},{"id":"2607.00890","title":"MultiSynt/MT: Trillion-Token Multi-Parallel Pre-Training Data Translated Across 36 Languages","created":"2026-07-01","cited_by_count":0,"area":null},{"id":"2607.04563","title":"Fidelity-Diversity Metrics for Text","created":"2026-07-06","cited_by_count":0,"area":null},{"id":"2607.22662","title":"CuraWeb: Joint Optimization of Quality, Redundancy, and Diversity for Web-Scale Pretraining Data","created":"2026-06-29","cited_by_count":0,"area":null},{"id":"2608.16390","title":"Counting Documents Is Not Counting Text: Unit Bias in Web-PDF Corpus Statistics","created":"2026-08-17","cited_by_count":0,"area":null},{"id":"2609.07699","title":"Fine PT-PT Web: A High-Quality 41 Billion Tokens Data Collection of the European Portuguese Web","created":"2026-09-07","cited_by_count":0,"area":null},{"id":"2609.16540","title":"On the Importance of Gating: Memorization vs. In-Context Learning in State Space Models","created":"2026-09-15","cited_by_count":0,"area":null},{"id":"2609.32484","title":"Towards Scalable Data Diversification for Language Model Pretraining via Leverage Score Sampling","created":"2026-09-26","cited_by_count":0,"area":null},{"id":"2609.34287","title":"ReScraper: Unified Scraping and Cleaning of Web Data for Effective LLM Pretraining","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2609.37891","title":"It's All Training: A Fully Synthetic Single-Stage Recipe for LLMs","created":"2026-09-29","cited_by_count":0,"area":null},{"id":"2610.10738","title":"Lossy Compressive Text Autoencoders","created":"2026-10-07","cited_by_count":0,"area":null},{"id":"2610.13735","title":"Orthros: Phase-Aware Heterogeneous Attention for Efficient Transformers","created":"2026-10-09","cited_by_count":0,"area":null},{"id":"2610.14709","title":"Vela: User Behavior as Supervision Sparks LLM Intelligence","created":"2026-10-10","cited_by_count":0,"area":null},{"id":"2610.17018","title":"LLM Evaluation is a Forecasting Problem","created":"2026-10-12","cited_by_count":0,"area":null},{"id":"2610.20086","title":"AFCS: Atomic-Fact-Conditioned Synthesis for Factual Knowledge Pretraining","created":"2026-10-14","cited_by_count":0,"area":null}],"next_cursor":null}