# To Copy or Not to Copy: Copying Is Easier to Induce Than Recall

arXiv: https://arxiv.org/abs/2601.12075 · HTML: https://mlsanity.com/paper/2601.12075/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation.
- Builds on 28 of its 32 references that are in the map.
- First posted on arXiv 17 Jan 2026.

## Placed in

- Retrieval-augmented generation / [RAG robustness and adaptive retrieval](https://mlsanity.com/rag/rag-robustness-and-adaptive-retrieval/)

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

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [REALM: Retrieval-Augmented Language Model Pre-Training](https://mlsanity.com/paper/2002.08909/) (2020)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [Language Models as Knowledge Bases?](https://mlsanity.com/paper/1909.01066/) (2019)
- [Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity](https://mlsanity.com/paper/2104.08786/) (2021)
- [When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories](https://mlsanity.com/paper/2212.10511/) (2022)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [How Much Knowledge Can You Pack Into the Parameters of a Language Model?](https://mlsanity.com/paper/2002.08910/) (2020)
- [In-Context Retrieval-Augmented Language Models](https://mlsanity.com/paper/2302.00083/) (2023)
- [Dissecting Recall of Factual Associations in Auto-Regressive Language Models](https://mlsanity.com/paper/2304.14767/) (2023)
- [Large Language Models Struggle to Learn Long-Tail Knowledge](https://mlsanity.com/paper/2211.08411/) (2022)
- [Measuring and Improving Consistency in Pretrained Language Models](https://mlsanity.com/paper/2102.01017/) (2021)
- [Simple Entity-Centric Questions Challenge Dense Retrievers](https://mlsanity.com/paper/2109.08535/) (2021)
- [Cutting Off the Head Ends the Conflict: A Mechanism for Interpreting and Mitigating Knowledge Conflicts in Language Models](https://mlsanity.com/paper/2402.18154/) (2024)
- [Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering](https://mlsanity.com/paper/2410.15999/) (2024)
- [Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals](https://mlsanity.com/paper/2402.11655/) (2024)
- [Serial Position Effects of Large Language Models](https://mlsanity.com/paper/2406.15981/) (2024)
- [Encoder-Decoder Gemma: Improving the Quality-Efficiency Trade-Off via Adaptation](https://mlsanity.com/paper/2504.06225/) (2025)
- [Is Your LLM Outdated? A Deep Look at Temporal Generalization](https://mlsanity.com/paper/2405.08460/) (2024)
- [Controllable Context Sensitivity and the Knob Behind It](https://mlsanity.com/paper/2411.07404/) (2024)
- [From RAGs to rich parameters: Probing how language models utilize external knowledge over parametric information for factual queries](https://mlsanity.com/paper/2406.12824/) (2024)
- [Order Matters: Investigate the Position Bias in Multi-constraint Instruction Following](https://mlsanity.com/paper/2502.17204/) (2025)
- [Deciphering the Interplay of Parametric and Non-parametric Memory in Retrieval-augmented Language Models](https://mlsanity.com/paper/2410.05162/) (2024)
- [Positional Biases Shift as Inputs Approach Context Window Limits](https://mlsanity.com/paper/2508.07479/) (2025)
- [Quantifying reliance on external information over parametric knowledge during Retrieval Augmented Generation (RAG) using mechanistic analysis](https://mlsanity.com/paper/2410.00857/) (2024)

## Built on by (most cited first, 1 of 1)

- [Plan Pointers and Record-Directive Form in Budgeted Verification of Inherited Agent Memory](https://mlsanity.com/paper/2609.03450/) (2026)

## Abstract (arXiv)

Language models used in retrieval-augmented settings must arbitrate between parametric knowledge stored in their weights and contextual information in the prompt. This work presents a mechanistic study of that choice by extracting an arbitration vector from model activations on a curated dataset designed to disentangle (i) irrelevant contexts that elicit parametric recall and (ii) relevant but false contexts that elicit copying. The vector is computed as the residual-stream centroid difference between these regimes across 27 relations, and is injected as an additive intervention at selected layers and token spans to steer behavior in two directions: CopyrightarrowRecall (suppressing context use) and RecallrightarrowCopy (inducing the model to copy any token from the context). Experiments on three architectures (decoder-only and encoder/decoder) and two open-domain QA benchmarks show consistent behavior shifts under moderate scaling while monitoring accuracy and fluency. Mechanistic analyses of attention routing, MLP contributions, and layer-wise probability trajectories reveal an asymmetry: inducing copying is an easy ``reactivation'' process that can be triggered at different locations in the input, while restoring recall is a ``suppression'' process that is more fragile and strongly tied to object-token interventions.