collaborators

6 papers

cs.CL2026

Understanding Fact Recall in Language Models: Why Two-Stage Training Encourages Memorization but Mixed Training Teaches Knowledge

Ying Zhang, Benjamin Heinzerling, Dongyuan Li +1

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 u…

cs.CL2025

How a Bilingual LM Becomes Bilingual: Tracing Internal Representations with Sparse Autoencoders

Tatsuro Inaba, Go Kamoda, Kentaro Inui +5

This study explores how bilingual language models develop complex internal representations. We employ sparse autoencoders to analyze internal representations of bilingual language…

cs.CL2025

Do LLMs Need to Think in One Language? Correlation between Latent Language and Task Performance

Shintaro Ozaki, Tatsuya Hiraoka, Hiroto Otake +8

Large Language Models (LLMs) are known to process information using a proficient internal language consistently, referred to as latent language, which may differ from the input or…

cs.CL2025

Weight-based Analysis of Detokenization in Language Models: Understanding the First Stage of Inference Without Inference

Go Kamoda, Benjamin Heinzerling, Tatsuro Inaba +3

According to the stages-of-inference hypothesis, early layers of language models map their subword-tokenized input, which does not necessarily correspond to a linguistically meanin…

cs.CL2025

The Geometry of Numerical Reasoning: Language Models Compare Numeric Properties in Linear Subspaces

Ahmed Oumar El-Shangiti, Tatsuya Hiraoka, Hilal AlQuabeh +2

This paper investigates whether large language models (LLMs) utilize numerical attributes encoded in a low-dimensional subspace of the embedding space when answering questions invo…

cs.CL2025

RECALL: Library-Like Behavior In Language Models is Enhanced by Self-Referencing Causal Cycles

Munachiso Nwadike, Zangir Iklassov, Toluwani Aremu +6

We introduce the concept of the self-referencing causal cycle (abbreviated RECALL) - a mechanism that enables large language models (LLMs) to bypass the limitations of unidirection…