language model training

Rethinking Data Mixture for Large Language Models: A Comprehensive Survey and New Perspectives EACL 2026 A Proactive Reliability Metric for Detecting Failures in Language Model Training EMNLP 2025 AutoMixer: Checkpoint Artifacts as Automatic Data Mixers ACL 2025 Balancing the Budget: Understanding Trade-offs Between Supervised and Preference-Based Finetuning ACL 2025 Developing and Utilizing a Large-Scale Cantonese Dataset for Multi-Tasking in Large Language Models EMNLP 2025 DIDS: Domain Impact-aware Data Sampling for Large Language Model Training EMNLP 2025 Empirical Evaluation of Loss Masking to Selectively Prevent Memorization ACL 2025 FastMCTS: A Simple Sampling Strategy for Data Synthesis ACL 2025 LDAdam: Adaptive Optimization from Low-Dimensional Gradient Statistics ICLR 2025 LLäMmlein: Transparent, Compact and Competitive German-Only Language Models from Scratch ACL 2025 MANTA: A Scalable Pipeline for Transmuting Massive Web Corpora into Instruction Datasets EMNLP 2025 Matina: A Large-Scale 73B Token Persian Text Corpus NAACL 2025 NOVER: Incentive Training for Language Models via Verifier-Free Reinforcement Learning EMNLP 2025 Phases of Uncertainty: Confidence–Calibration Dynamics in Language Model Training EMNLP 2025 Rationales Are Not Silver Bullets: Measuring the Impact of Rationales on Model Performance and Reliability ACL 2025 SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training ICLR 2025 Studying the Role of Input-Neighbor Overlap in Retrieval-Augmented Language Models Training Efficiency EMNLP 2025 Towards Open Foundation Language Model and Corpus for Macedonian: A Low-Resource Language ACL 2025 T-REG: Preference Optimization with Token-Level Reward Regularization ACL 2025 Understanding Reference Policies in Direct Preference Optimization NAACL 2025 VerIF: Verification Engineering for Reinforcement Learning in Instruction Following EMNLP 2025 What did you say? Generating Child-Directed Speech Questions to Train LLMs EMNLP 2025 β-DPO: Direct Preference Optimization with Dynamic β NeurIPS 2024 Balanced Data Sampling for Language Model Training with Clustering ACL 2024 CSLM: A Framework for Question Answering Dataset Generation through Collaborative Small Language Models EMNLP 2024