scaling law

Beyond Scaling: Measuring and Predicting the Upper Bound of Knowledge Retention in Language Model Pre-Training ACL 2026 Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models ACL 2026 Inference Scaling Law for Retrieval Augmented Generation AAAI 2026 Optimal Expert-Attention Allocation in Mixture-of-Experts: A Scalable Law for Dynamic Model Design ACL 2026 Optimal L2 Regularization in High-dimensional Continual Linear Regression ALT 2026 Predicting Emergent Tool Use in LLMs Before It Emerges: A Proxy Perspective AAAI 2026 Scaling Behaviors of LLM Reinforcement Learning Post-Training: An Empirical Study in Mathematical Reasoning ACL 2026 Scaling Law for Large Wireless Models AAAI 2026 Scaling Law for Multimodal Large Language Model Supervised Fine-Tuning ACL 2026 Scaling Laws for Code: A More Data-Hungry Regime ACL 2026 Scaling Laws for Conditional Emergence of Multilingual Image Captioning via Generalization from Translation AAAI 2026 Scaling View Synthesis Transformers CVPR 2026 The Retrieval Bottleneck: Scaling Laws for Reinforcement Learning in RAG ACL 2026 Adaptive Data Optimization: Dynamic Sample Selection with Scaling Laws ICLR 2025 A Hitchhiker’s Guide to Scaling Law Estimation ICML 2025 An Empirical Study of Autoregressive Pre-training from Videos ICCV 2025 A Survey of Post-Training Scaling in Large Language Models ACL 2025 Bayesian Neural Scaling Law Extrapolation with Prior-Data Fitted Networks ICML 2025 Beyond Text Compression: Evaluating Tokenizers Across Scales ACL 2025 Breaking Neural Network Scaling Laws with Modularity ICLR 2025 Capability Salience Vector: Fine-grained Alignment of Loss and Capabilities for Downstream Task Scaling Law ACL 2025 Compute-Optimal LLMs Provably Generalize Better with Scale ICLR 2025 Compute Optimal Scaling of Skills: Knowledge vs Reasoning ACL 2025 Data Mixing Laws: Optimizing Data Mixtures by Predicting Language Modeling Performance ICLR 2025 Data Mixing Optimization for Supervised Fine-Tuning of Large Language Models ICML 2025