← Search

Koshi Eguchi

2 accepted papers

2026

Extending the Context of Pretrained LLMs by Dropping Their Positional Embedding

ICLR 2026poster

So far, expensive finetuning beyond the pretraining sequence length has been a prerequisite to effectively extend the context of language models (LM). In this work, we break this key bottleneck by ***Dro**pping the **P**ositional **E**mbeddings of LMs after training (DroPE)*. Our simple method is mo…

Cited by 17SourcecodeScholar
2026

Steering at the Source: Style Modulation Heads for Robust Persona Control

ICML 2026poster

Activation steering offers a computationally efficient mechanism for controlling Large Language Models (LLMs) without fine-tuning. While effectively controlling target traits (e.g., persona), coherency degradation remains a major obstacle to safety and practical deployment. We hypothesize that this …

Cited by 0SourceScholar