一作ACL 2026Workshop· Natural Language Processing
One Task Vector is not Enough: A Large-Scale Study for In-Context Learning
Pavel Tikhonov · Ivan Oseledets · Elena Tutubalina
摘要
In-context learning (ICL) enables Large Language Models (LLMs) to adapt to new tasks using few examples, with task vectors, defined as specific hidden state activations hypothesized to encode task information. Existing studies are limited by small-scale benchmarks, restricting comprehensive analy...