Benjamin Heinzerling
3 papers in the PaperMetrix corpus
Papers by this author
-
Test-time Augmentation for Factual Probing
2023
Factual probing is a method that uses prompts to test if a language model “knows” certain world knowledge facts. A problem in factual probing is that small changes to the prompt can lead to large …
-
Understanding Fact Recall in Language Models: Why Two-Stage Training Encourages Memorization but Mixed Training Teaches Knowledge
2025 · arXiv (Cornell University)
While fine-tuning is the standard for injecting factual knowledge into large language models (LLMs), the mechanisms enabling reliable fact recall via unseen queries remain poorly understood. Common two-stage training strategies, which sequentially train on fact …
-
BPEmb: Tokenization-free Pre-trained Subword Embeddings in 275 Languages
2017 · arXiv (Cornell University)
We present BPEmb, a collection of pre-trained subword unit embeddings in 275 languages, based on Byte-Pair Encoding (BPE). In an evaluation using fine-grained entity typing as testbed, BPEmb performs competitively, and for some languages bet- …