# LinearARD: Linear-Memory Attention Distillation for RoPE Restoration

arXiv: https://arxiv.org/abs/2604.00004 · HTML: https://mlsanity.com/paper/2604.00004/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 32 of its 33 references that are in the map.
- First posted on arXiv 9 Mar 2026.

## Builds on (most cited first, 30 of 32)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering](https://mlsanity.com/paper/1809.02789/) (2018)
- [Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection](https://mlsanity.com/paper/2310.11511/) (2023)
- [MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers](https://mlsanity.com/paper/2002.10957/) (2020)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [TinyBERT: Distilling BERT for Natural Language Understanding](https://mlsanity.com/paper/1909.10351/) (2019)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [Extending Context Window of Large Language Models via Positional Interpolation](https://mlsanity.com/paper/2306.15595/) (2023)
- [MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms](https://mlsanity.com/paper/1905.13319/) (2019)
- [YaRN: Efficient Context Window Extension of Large Language Models](https://mlsanity.com/paper/2309.00071/) (2023)
- [MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices](https://mlsanity.com/paper/2004.02984/) (2020)
- [Ring Attention with Blockwise Transformers for Near-Infinite Context](https://mlsanity.com/paper/2310.01889/) (2023)
- [LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens](https://mlsanity.com/paper/2402.13753/) (2024)
- [MiniLMv2: Multi-Head Self-Attention Relation Distillation for Compressing Pretrained Transformers](https://mlsanity.com/paper/2012.15828/) (2020)
- [Continual Pre-training of Language Models](https://mlsanity.com/paper/2302.03241/) (2023)
- [PoSE: Efficient Context Window Extension of LLMs via Positional Skip-wise Training](https://mlsanity.com/paper/2309.10400/) (2023)

## Abstract (arXiv)

The extension of context windows in Large Language Models is typically facilitated by scaling positional encodings followed by lightweight Continual Pre-Training (CPT). While effective for processing long sequences, this paradigm often disrupts original model capabilities, leading to performance degradation on standard short-text benchmarks. We propose LinearARD, a self-distillation method that restores Rotary Position Embeddings (RoPE)-scaled students through attention-structure consistency with a frozen native-RoPE teacher. Rather than matching opaque hidden states, LinearARD aligns the row-wise distributions of dense Q/Q, K/K, and V/V self-relation matrices to directly supervise attention dynamics. To overcome the quadratic memory bottleneck of n × n relation maps, we introduce a linear-memory kernel. This kernel leverages per-token log-sum-exp statistics and fuses logit recomputation into the backward pass to compute exact Kullback-Leibler divergence and gradients. On LLaMA2-7B extended from 4K to 32K, LinearARD recovers 98.3% of the short-text performance of state-of-the-art baselines while surpassing them on long-context benchmarks. Notably, our method achieves these results using only 4.25M training tokens compared to the 256M tokens required by LongReD and CPT. Our code is available at https://github.com/gracefulning/LinearARD.