NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
49 results for “keyword:"language model pretraining"” in ICLR 2027. 5 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Don't Discard, Repair: A Repair-First Paradigm for Recovering Training Utility from Imperfect Pre-training Corpora▼4.3k2.00
[PDF]The COUNT: Counting N-grams Unlocks Scaling on Host Memory and Training-Free Adaptation▼4.5k2.67
[PDF]ChromaMuon: Spectral Optimizer Diagnostics and Design with Retained Newton–Schulz Coordinates▼1.2k1.67
[PDF]Learning Less Is More: Premature Upper-Layer Attention Specialization Hurts Language Model Pretraining▼1.2k1.67
[PDF]Token Mass, Not Geometry: A Confound in Embedding-Based Data Selection for Language Model Pretraining▼1.3k1.75
[PDF]One Marginal Is Enough: What Doubly Stochastic Routing Buys Hyper-Connections, and What It Costs▲3.0k2.33
[PDF]Beyond Fact Frequency: Controlled Evidence on Expression Allocation and Factual Callability▼1.2k1.67
[PDF]Future Data Allocation Shapes Repetition-History Persistence in Language Model Pretraining▲5.7k2.00