In-context learning creates task vectors
arxiv.org · 6,018 words · saved by 1 readers
N/A
In-Context Learning Creates Task Vectors Roee Hendel Mor Geva Amir Globerson Tel Aviv University Google DeepMind Tel Aviv University, Google roee.hendel@mail.tau.ac.il pipek@google.com gamir@tauex.tau.ac.il Abstract Large Language Model…
related reading
- In-context Learning and Induction Headstransformer-circuits.pub
- NL.pdfabehrouz.github.io
- How does in-context learning work? A framework for understanding the differences from traditional supervised learning | SAIL Blogai.stanford.edu
- A mechanism for solving relational tasks in transformer language modelsarxiv.org
- [2005.14165] Language Models are Few-Shot Learnersarxiv.org
- ICLR: In-Context Learning of Representationsarxiv.org
- Language Models can Solve Computer Tasksarxiv.org
- [2511.08579] Training Language Models to Explain Their Own Computationsarxiv.org
- Are Emergent Abilities in Large Language Models just In-Context Learning?arxiv.org
- Learning without training: The implicit dynamics of in-context learningarxiv.org
- [2005.11401] Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasksarxiv.org
- Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?arxiv.org