# Information-Preserving Reformulation of Reasoning Traces for Antidistillation

arXiv: https://arxiv.org/abs/2510.11545 · HTML: https://mlsanity.com/paper/2510.11545/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning.
- Builds on 5 of its 32 references that are in the map.
- First posted on arXiv 13 Oct 2025.

## Placed in

- LLM reasoning / [Prompting and chain-of-thought reasoning](https://mlsanity.com/lr/prompting-and-chain-of-thought-reasoning/)

## Builds on (most cited first, 5 of 5)

- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [Antidistillation Sampling](https://mlsanity.com/paper/2504.13146/) (2025)
- [DOGe: Defensive Output Generation for LLM Protection Against Knowledge Distillation](https://mlsanity.com/paper/2505.19504/) (2025)

## Built on by (most cited first, 8 of 8)

- [Protecting Language Models Against Unauthorized Distillation through Trace Rewriting](https://mlsanity.com/paper/2602.15143/) (2026)
- [RedAct: Redacting Agent Capability Traces for Procedural Skill Protection](https://mlsanity.com/paper/2606.10813/) (2026)
- [Asking Back: Interaction-Layer Antidistillation Watermarks](https://mlsanity.com/paper/2605.16462/) (2026)
- [Safeguarding Text-to-Image Generative Models Against Unauthorized Knowledge Distillation](https://mlsanity.com/paper/2605.22060/) (2026)
- [What Does It Mean to Break a Distillation Defense?](https://mlsanity.com/paper/2606.25059/) (2026)
- [Metacognitive Behavioral Tuning of Large Language Models for Multi-Hop Question Answering](https://mlsanity.com/paper/2602.22508/) (2026)
- [Hiding in Plain Sight: Detectability-Aware Antidistillation of Reasoning Models](https://mlsanity.com/paper/2604.23238/) (2026)
- [The Distillation Game: Adaptive Evaluations & Efficient Defenses](https://mlsanity.com/paper/2605.22737/) (2026)

## Abstract (arXiv)

Recent advances in Large Language Models (LLMs) show that extending the length of reasoning chains significantly improves performance on complex tasks. While revealing these reasoning traces helps users better follow, verify, and learn from the model's problem-solving process, it also makes them highly vulnerable to unauthorized distillation. To mitigate this risk, proprietary model providers often adopt aggressive protection strategies, such as replacing detailed reasoning with brief summaries, which deprive users of valuable intermediate information. To address this trade-off, we propose PART, an information-preserving antidistillation reformulation of reasoning traces. Motivated by the difference between how humans understand reasoning traces and how LLMs exploit them for supervised fine-tuning, we design a simple but effective two-step reformulation: removing self-talk behaviors and reordering sub-conclusions. A small auxiliary model is trained to perform this reformulation, incurring minimal computational overhead. Extensive experiments demonstrate that PART consistently disrupts distillation across student models of different sizes and types on various reasoning benchmarks. For instance, when training on reformulated traces, even the performance of a large 32B student model decreases from 54.17 to 46.88 on AIME 2024, corresponding to a 13.5% degradation.