All or None: Identifiable Linear Properties of Next-token Predictors in Language Modeling

30 October 2024

Papers citing "All or None: Identifiable Linear Properties of Next-token Predictors in Language Modeling"

3 / 3 papers shown

Title
An Empirically Grounded Identifiability Theory Will Accelerate Self-Supervised Learning Research Patrik Reizinger Randall Balestriero David Klindt Wieland Brendel 31 0 0 17 Apr 2025
I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data? Yuhang Liu Dong Gong Erdun Gao Zhen Zhang Biwei Huang Mingming Gong Anton van den Hengel Javen Qinfeng Shi J. Shi 55 0 0 12 Mar 2025
What is causal about causal models and representations? Frederik Hytting Jørgensen Luigi Gresele S. Weichwald CML 97 0 0 31 Jan 2025