# Just-in-time and distributed task representations in language models

arXiv: https://arxiv.org/abs/2509.04466 · HTML: https://mlsanity.com/paper/2509.04466/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 31 references that are in the map.
- First posted on arXiv 28 Aug 2025.

## Builds on (most cited first, 28 of 28)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?](https://mlsanity.com/paper/2202.12837/) (2022)
- [An Explanation of In-context Learning as Implicit Bayesian Inference](https://mlsanity.com/paper/2111.02080/) (2021)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Transformers learn in-context by gradient descent](https://mlsanity.com/paper/2212.07677/) (2022)
- [What learning algorithm is in-context learning? Investigations with linear models](https://mlsanity.com/paper/2211.15661/) (2022)
- [In-Context Learning Creates Task Vectors](https://mlsanity.com/paper/2310.15916/) (2023)
- [Data Distributional Properties Drive Emergent In-Context Learning in Transformers](https://mlsanity.com/paper/2205.05055/) (2022)
- [Function Vectors in Large Language Models](https://mlsanity.com/paper/2310.15213/) (2023)
- [Many-Shot In-Context Learning](https://mlsanity.com/paper/2404.11018/) (2024)
- [The Evolution of Statistical Induction Heads: In-Context Learning Markov Chains](https://mlsanity.com/paper/2402.11004/) (2024)
- [Which Attention Heads Matter for In-Context Learning?](https://mlsanity.com/paper/2502.14010/) (2025)
- [Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learning](https://mlsanity.com/paper/2406.15334/) (2024)
- [The broader spectrum of in-context learning](https://mlsanity.com/paper/2412.03782/) (2024)
- [Parallel Structures in Pre-training Data Yield In-Context Learning](https://mlsanity.com/paper/2402.12530/) (2024)
- [Task Vectors in In-Context Learning: Emergence, Formation, and Benefit](https://mlsanity.com/paper/2501.09240/) (2025)
- [Revisiting In-context Learning Inference Circuit in Large Language Models](https://mlsanity.com/paper/2410.04468/) (2024)
- [Strategy Coopetition Explains the Emergence and Transience of In-Context Learning](https://mlsanity.com/paper/2503.05631/) (2025)
- [Vision-Language Models Create Cross-Modal Task Representations](https://mlsanity.com/paper/2410.22330/) (2024)
- [Everything Everywhere All at Once: LLMs can In-Context Learn Multiple Tasks in Superposition](https://mlsanity.com/paper/2410.05603/) (2024)
- [Implicit In-context Learning](https://mlsanity.com/paper/2405.14660/) (2024)
- [Learning Task Representations from In-Context Learning](https://mlsanity.com/paper/2502.05390/) (2025)
- [Do different prompting methods yield a common task representation in language models?](https://mlsanity.com/paper/2505.12075/) (2025)
- [Contextualize-then-Aggregate: Circuits for In-Context Learning in Gemma-2 2B](https://mlsanity.com/paper/2504.00132/) (2025)
- [Understanding Task Vectors in In-Context Learning: Emergence, Functionality, and Limitations](https://mlsanity.com/paper/2506.09048/) (2025)
- [One Task Vector is not Enough: A Large-Scale Study for In-Context Learning](https://mlsanity.com/paper/2505.23911/) (2025)

## Built on by (most cited first, 9 of 9)

- [Monitoring Emergent Reward Hacking During Generation via Internal Activations](https://mlsanity.com/paper/2603.04069/) (2026)
- [Linear representations in language models can change dramatically over a conversation](https://mlsanity.com/paper/2601.20834/) (2026)
- [Context Structure Reshapes the Representational Geometry of Language Models](https://mlsanity.com/paper/2601.22364/) (2026)
- [Convergent World Representations and Divergent Tasks](https://mlsanity.com/paper/2602.00533/) (2026)
- [Decomposing how prompting steers behavior](https://mlsanity.com/paper/2606.03093/) (2026)
- [Patches of Nonlinearity: Instruction Vectors in Large Language Models](https://mlsanity.com/paper/2602.07930/) (2026)
- [Radical AI Interpretability](https://mlsanity.com/paper/2606.26523/) (2026)
- [Distributed Sparse Interventions in Language Models](https://mlsanity.com/paper/2607.07128/) (2026)
- [One mechanism for many mental spaces: a shared router over a value slot in language models](https://mlsanity.com/paper/2607.10248/) (2026)

## Abstract (arXiv)

Many of language models' impressive capabilities originate from their in-context learning: based on instructions or examples, they can infer and perform new tasks without weight updates. In this work, we investigate when representations for new tasks are formed in language models, and how these representations change over the course of context. We study two different task representations: those that are ''transferrable'' -- vector representations that can transfer task contexts to another model instance, even without the full prompt -- and simpler representations of high-level task categories. We show that transferrable task representations evolve in non-monotonic and sporadic ways, while task identity representations persist throughout the context. Specifically, transferrable task representations exhibit a two-fold locality. They successfully condense evidence when more examples are provided in the context. But this evidence accrual process exhibits strong temporal locality along the sequence dimension, coming online only at certain tokens -- despite task identity being reliably decodable throughout the context. In some cases, transferrable task representations also show semantic locality, capturing a small task ''scope'' such as an independent subtask. Language models thus represent new tasks on the fly through both an inert, sustained sensitivity to the task and an active, just-in-time representation to support inference.