conftrace_

probing framework

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

probing test (1) model representation (1) bias extraction (1) token language alignment (1) abstraction capability (1) two-stage training process (1) abstract concept (1) linear separability (1) vocabulary sharing (1) multilingual representation (1)

Papers

Language Directions in Multilingual LLMs: A Layer-wise Diagnostic Study of Token Alignment and Pretraining Imprint ACL 2026 Does Deep Learning Learn to Abstract? A Systematic Probing Framework ICLR 2023 Debiasing Methods in Natural Language Understanding Make Bias More Accessible EMNLP 2021