works on

From the 1 of 10 linked papers with an AI index.

collaborators

10 papers

cs.CL2026

Relation Geometry in Semantic Space of Language Models

Zhihan Cao, Hiroaki Yamada, Simone Teufel +4

The paper investigates how different semantic relations are reflected in the geometric structure of word embedding spaces produced by various language models, examining region clus…

cs.CL2026

Sycophancy Hides Linearly in the Attention Heads

Rifo Genadi, Munachiso Nwadike, Nurdaulet Mukhituly +3

We find that correct-to-incorrect sycophancy signals are most linearly separable within multi-head attention activations. Motivated by the linear representation hypothesis, we trai…

cs.CL2025

Understanding and Controlling Repetition Neurons and Induction Heads in In-Context Learning

Nhi Hoai Doan, Tatsuya Hiraoka, Kentaro Inui

This paper investigates the relationship between large language models' (LLMs) ability to recognize repetitive input patterns and their performance on in-context learning (ICL). In…

cs.CL2025

Augmenting Dialog with Think-Aloud Utterances for Modeling Individual Personality Traits by LLM

Seiya Ishikura, Hiroaki Yamada, Tatsuya Hiraoka +1

This study proposes augmenting dialog data with think-aloud utterances (TAUs) for modeling individual personalities in text chat by LLM. TAU is a verbalization of a speaker's thoug…

cs.CL2025

Spelling-out is not Straightforward: LLMs' Capability of Tokenization from Token to Characters

Tatsuya Hiraoka, Kentaro Inui

Large language models (LLMs) can spell out tokens character by character with high accuracy, yet they struggle with more complex character-level tasks, such as identifying composit…

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…