{"total":160,"rows":[{"id":"2501.12599","title":"Kimi k1.5: Scaling Reinforcement Learning with LLMs","created":"2025-01-22","cited_by_count":1210,"area":"lr"},{"id":"2508.06471","title":"GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models","created":"2025-08-08","cited_by_count":467,"area":"lr"},{"id":"2504.07491","title":"Kimi-VL Technical Report","created":"2025-04-10","cited_by_count":388,"area":"lr"},{"id":"2506.04178","title":"OpenThoughts: Data Recipes for Reasoning Models","created":"2025-06-04","cited_by_count":304,"area":"lr"},{"id":"2510.25741","title":"Scaling Latent Reasoning via Looped Language Models","created":"2025-10-29","cited_by_count":165,"area":"lr"},{"id":"2510.26583","title":"Emu3.5: Native Multimodal Models are World Learners","created":"2025-10-30","cited_by_count":132,"area":"wm"},{"id":"2508.14444","title":"NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model","created":"2025-08-20","cited_by_count":108,"area":"lr"},{"id":"2512.20848","title":"Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning","created":"2025-12-23","cited_by_count":104,"area":"lr"},{"id":"2504.03624","title":"Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models","created":"2025-04-04","cited_by_count":100,"area":"lr"},{"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":"2508.15763","title":"Intern-S1: A Scientific Multimodal Foundation Model","created":"2025-08-21","cited_by_count":53,"area":"moe"},{"id":"2504.02807","title":"MegaMath: Pushing the Limits of Open Math Corpora","created":"2025-04-03","cited_by_count":51,"area":"lr"},{"id":"2511.03276","title":"Diffusion Language Models are Super Data Learners","created":"2025-11-05","cited_by_count":48,"area":"diffusion"},{"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":"2508.15096","title":"Nemotron-CC-Math: A 133 Billion-Token-Scale High Quality Math Pretraining Dataset","created":"2025-08-20","cited_by_count":38,"area":"lr"},{"id":"2512.10858","title":"Scaling Behavior of Discrete Diffusion Language Models","created":"2025-12-11","cited_by_count":35,"area":"diffusion"},{"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":"2510.03280","title":"Training Optimal Large Diffusion Language Models","created":"2025-09-28","cited_by_count":25,"area":"diffusion"},{"id":"2502.11191","title":"Primus: A Pioneering Collection of Open-Source Datasets for Cybersecurity LLM Training","created":"2025-02-16","cited_by_count":23,"area":"distill"},{"id":"2505.02881","title":"Rewriting Pre-Training Data Boosts LLM Performance in Math and Code","created":"2025-05-05","cited_by_count":21,"area":"lr"},{"id":"2508.10975","title":"BeyondWeb: Lessons from Scaling Synthetic Data for Trillion-scale Pretraining","created":"2025-08-14","cited_by_count":21,"area":"synth"},{"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":"2512.24617","title":"Dynamic Large Concept Models: Latent Reasoning in an Adaptive Semantic Space","created":"2025-12-31","cited_by_count":17,"area":"lr"},{"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":"2510.24695","title":"AgentFrontier: Expanding the Capability Frontier of LLM Agents with ZPD-Guided Data Synthesis","created":"2025-10-28","cited_by_count":14,"area":"agents"},{"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":"2506.22403","title":"HyperCLOVA X THINK Technical Report","created":"2025-06-27","cited_by_count":7,"area":"lr"},{"id":"2509.26520","title":"Training Matryoshka Mixture-of-Experts for Elastic Inference-Time Expert Utilization","created":"2025-09-30","cited_by_count":7,"area":"moe"},{"id":"2510.05781","title":"Mixture of Neuron Experts","created":"2025-10-07","cited_by_count":7,"area":"moe"},{"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":"2604.24432","title":"Kwai Summary Attention Technical Report","created":"2026-04-27","cited_by_count":7,"area":"lr"},{"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":"2505.00022","title":"Aleph-Alpha-GermanWeb: Improving German-language LLM pre-training with model-based data curation and synthetic data generation","created":"2025-04-24","cited_by_count":6,"area":"synth"},{"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":"2509.15248","title":"Synthetic bootstrapped pretraining","created":"2025-09-17","cited_by_count":6,"area":"synth"},{"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":"2602.21472","title":"The Design Space of Tri-Modal Masked Diffusion Models","created":"2026-02-25","cited_by_count":6,"area":"diffusion"},{"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":"2505.07293","title":"AttentionInfluence: Adopting Attention Head Influence for Weak-to-Strong Pretraining Data Selection","created":"2025-05-12","cited_by_count":5,"area":"lr"},{"id":"2507.15640","title":"Data Mixing Agent: Learning to Re-weight Domains for Continual Pre-training","created":"2025-07-21","cited_by_count":5,"area":"agents"},{"id":"2507.20984","title":"SmallThinker: A Family of Efficient Large Language Models Natively Trained for Local Deployment","created":"2025-07-28","cited_by_count":5,"area":"lr"},{"id":"2601.09066","title":"Mi:dm 2.0 Korea-centric Bilingual Language Models","created":"2026-01-14","cited_by_count":5,"area":null},{"id":"2605.20948","title":"Memory Grafting: Scaling Language Model Pre-training via Offline Conditional Memory","created":"2026-05-20","cited_by_count":5,"area":"moe"},{"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":"2506.07463","title":"CCI4.0: A Bilingual Pretraining Dataset for Enhancing Reasoning in Large Language Models","created":"2025-06-09","cited_by_count":4,"area":"lr"},{"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":"2510.09426","title":"KORMo: Korean Open Reasoning Model for Everyone","created":"2025-10-10","cited_by_count":4,"area":"lr"},{"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":"2602.19548","title":"Beyond a Single Extractor: Re-thinking HTML-to-Text Extraction for LLM Pretraining","created":"2026-02-23","cited_by_count":4,"area":"lr"},{"id":"2603.00054","title":"Expert Divergence Learning for MoE-based Language Models","created":"2026-02-10","cited_by_count":4,"area":"moe"},{"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":"2509.25531","title":"MixtureVitae: Open Web-Scale Pretraining Dataset With High Quality Instruction and Reasoning Data Built from Permissive-First Text Sources","created":"2025-09-29","cited_by_count":3,"area":"lr"},{"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":"2604.01622","title":"Expert-Choice Routing Enables Adaptive Computation in Diffusion Language Models","created":"2026-04-02","cited_by_count":3,"area":"diffusion"},{"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":"2504.15527","title":"Compass-V2 Technical Report","created":"2025-04-22","cited_by_count":2,"area":"moe"},{"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":"2511.10338","title":"BhashaKritika: Building Synthetic Pretraining Data at Scale for Indic Languages","created":"2025-11-13","cited_by_count":2,"area":"synth"},{"id":"2512.07612","title":"PCMind-2.1-Kaiyuan-2B Technical Report","created":"2025-12-08","cited_by_count":2,"area":null},{"id":"2512.16248","title":"Sigma-MoE-Tiny Technical Report","created":"2025-12-18","cited_by_count":2,"area":"moe"},{"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":"2605.11405","title":"20/20 Vision Language Models: A Prescription for Better VLMs through Data Curation Alone","created":"2026-05-12","cited_by_count":2,"area":"mllm"}],"next_cursor":"100"}