paper / tallinzen / May 20 / failedSpatial homogeneity of superconducting order parameter in NbN films grown by atomic layer deposition
paper / tallinzen / May 19 / failedAlways Learning, Always Mixing: Efficient and Simple Data Mixing All The Time
paper / tallinzen / May 19 / failedWhy are language models less surprised than humans? Testing the Parse Multiplicity Mismatch Hypothesis
paper / tallinzen / May 14 / failedLanguage Models Struggle to Use Representations Learned In-Context
paper / tallinzen / May 14 / failedDeconstructing sentence disambiguation by joint latent modeling of reading paradigms: LLM surprisal is not enough
paper / tallinzen / May 14 / failedBabyLM Turns 4 and Goes Multilingual: Call for Papers for the 2026 BabyLM Workshop