{"total":48,"rows":[{"id":"2507.12466","title":"Language Models Improve When Pretraining Data Matches Target Tasks","created":"2025-07-16","cited_by_count":19,"area":null},{"id":"2508.17677","title":"TiKMiX: Take Data Influence into Dynamic Mixture for Language Model Pre-training","created":"2025-08-25","cited_by_count":6,"area":null},{"id":"2509.23500","title":"Beyond Outliers: A Study of Optimizers Under Quantization","created":"2025-09-27","cited_by_count":6,"area":null},{"id":"2602.07712","title":"Towards Robust Scaling Laws for Optimizers","created":"2026-02-07","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":"2505.19800","title":"MOLE: Metadata Extraction and Validation in Scientific Papers Using LLMs","created":"2025-05-26","cited_by_count":5,"area":null},{"id":"2606.27019","title":"MinGram: A Minimalist Unigram Tokenizer with High Compression and Competitive Morphological Alignment","created":"2026-06-25","cited_by_count":5,"area":null},{"id":"2505.20380","title":"GRAPE: Optimize Data Mixture for Group Robust Multi-target Adaptive Pretraining","created":"2025-05-26","cited_by_count":4,"area":null},{"id":"2605.22821","title":"Tokenisation via Convex Relaxations","created":"2026-05-21","cited_by_count":4,"area":null},{"id":"2608.11612","title":"Dion3: Full-Stack Orthogonal Updates","created":"2026-08-12","cited_by_count":4,"area":null},{"id":"2512.04555","title":"ADAPT: Learning Task Mixtures for Budget-Constrained Instruction Tuning","created":"2025-12-04","cited_by_count":3,"area":null},{"id":"2602.04289","title":"Proxy Compression for Language Modeling","created":"2026-02-04","cited_by_count":3,"area":null},{"id":"2505.23878","title":"AC-ODM: Actor--Critic Online Data Mixing for Sample-Efficient LLM Pretraining","created":"2025-05-29","cited_by_count":2,"area":null},{"id":"2509.25850","title":"RL-Guided Data Selection for Language Model Finetuning","created":"2025-09-30","cited_by_count":2,"area":null},{"id":"2512.15306","title":"LLMQ: Efficient Lower-Precision Pretraining for Consumer GPUs","created":"2025-12-17","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":"2605.18528","title":"Tight Complexity of Muon under Heavy-Tailed Noise","created":"2026-05-18","cited_by_count":2,"area":null},{"id":"2605.22705","title":"Tokenization with Split Trees","created":"2026-05-21","cited_by_count":2,"area":null},{"id":"2606.11499","title":"Hubs or Fringes: Pretraining Data Selection via Web Graph Centrality","created":"2026-06-09","cited_by_count":2,"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":"2605.17842","title":"SNLP: Layer-Parallel Inference via Structured Newton Corrections","created":"2026-05-18","cited_by_count":1,"area":null},{"id":"2605.20798","title":"Most Transformer Modifications Still Do Not Transfer at 1-3B: A 2020-2026 Update to Narang et al. (2021) with Downstream Evaluation and a Noise Floor","created":"2026-05-20","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":"2606.02780","title":"Do Value Vectors in Deep Layers Need Context from the Residual Stream?","created":"2026-06-01","cited_by_count":1,"area":null},{"id":"2606.03773","title":"KletterMix: Climbing Toward High-Quality German Pretraining Data - The Full Report","created":"2026-06-02","cited_by_count":1,"area":null},{"id":"2606.08167","title":"Explaining Data Mixing Scaling Laws","created":"2026-06-06","cited_by_count":1,"area":null},{"id":"2606.26050","title":"Natural Ungrokking: Asymmetric Control of Which Rules Survive Pretraining","created":"2026-06-24","cited_by_count":1,"area":null},{"id":"2608.05806","title":"Hierarchical Latent Prediction for Language Models","created":"2026-08-06","cited_by_count":1,"area":null},{"id":"2608.06283","title":"The Tamed Subgradient Unadjusted Langevin Algorithm beyond Convexity","created":"2026-08-06","cited_by_count":1,"area":null},{"id":"2609.03807","title":"Almost Free State Prediction Separation","created":"2026-09-03","cited_by_count":1,"area":null},{"id":"2602.11615","title":"SkillRater: Untangling Capabilities in Multimodal Data","created":"2026-02-12","cited_by_count":0,"area":null},{"id":"2604.26170","title":"EvoSelect: Data-Efficient LLM Evolution for Targeted Task Adaptation","created":"2026-04-28","cited_by_count":0,"area":null},{"id":"2606.15216","title":"Spokes: Optimizing for Diverse Pretraining Data Selection","created":"2026-06-13","cited_by_count":0,"area":null},{"id":"2607.02266","title":"HERMES: A Multi-Granularity Labeling Substrate for Pre-training Data Mixtures","created":"2026-07-02","cited_by_count":0,"area":null},{"id":"2607.04819","title":"Layer-Parallel Inference Reduces Encrypted Nonlinear Depth in Transformers","created":"2026-07-06","cited_by_count":0,"area":null},{"id":"2608.08847","title":"Explicit Boundary Markers for Subword Vocabularies","created":"2026-08-09","cited_by_count":0,"area":null},{"id":"2608.17120","title":"Children, but not language models, show accelerating returns in word learning","created":"2026-08-17","cited_by_count":0,"area":null},{"id":"2608.26973","title":"Squeezing More from Limited Data with Recursive Transformers","created":"2026-08-27","cited_by_count":0,"area":null},{"id":"2609.01573","title":"Scaling Near-Optimal SFT-RL Annotation Budget Allocation from Small to Large LLMs","created":"2026-09-01","cited_by_count":0,"area":null},{"id":"2609.07129","title":"Line-Coupled Language Model","created":"2026-09-07","cited_by_count":0,"area":null},{"id":"2609.13677","title":"Nonsmooth Optimization via Orthogonalized Momentum","created":"2026-09-12","cited_by_count":0,"area":null},{"id":"2609.20423","title":"WeVisDoc: From Coverage to Capability for Robust End-to-End Document Parsing","created":"2026-09-17","cited_by_count":0,"area":null},{"id":"2609.32759","title":"The Extender: A Log-Structured Transformer","created":"2026-09-26","cited_by_count":0,"area":null},{"id":"2609.34174","title":"GradLev: Token-Parallel Test-Time Training Via Costate Prediction","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2609.39114","title":"The Row Normalization Puzzle in Muon","created":"2026-09-30","cited_by_count":0,"area":null},{"id":"2610.05597","title":"More Than Words: Compositional Tokenization for Efficient Language Models","created":"2026-10-04","cited_by_count":0,"area":null},{"id":"2610.05978","title":"Byte Language Models: Scaling, Emergent Abstractions, and Information Allocation","created":"2026-10-05","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}],"next_cursor":null}