# Analyzing and Internalizing Complex Policy Documents for LLM Agents

arXiv: https://arxiv.org/abs/2510.11588 · HTML: https://mlsanity.com/paper/2510.11588/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 43 of its 52 references that are in the map.
- First posted on arXiv 13 Oct 2025.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 30 of 43)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [From Local to Global: A Graph RAG Approach to Query-Focused Summarization](https://mlsanity.com/paper/2404.16130/) (2024)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [Instruction-Following Evaluation for Large Language Models](https://mlsanity.com/paper/2311.07911/) (2023)
- [Learning to Compress Prompts with Gist Tokens](https://mlsanity.com/paper/2304.08467/) (2023)
- [Adapting Language Models to Compress Contexts](https://mlsanity.com/paper/2305.14788/) (2023)
- [K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapters](https://mlsanity.com/paper/2002.01808/) (2020)
- [Deliberative Alignment: Reasoning Enables Safer Language Models](https://mlsanity.com/paper/2412.16339/) (2024)
- [Evaluating Large Language Models at Evaluating Instruction Following](https://mlsanity.com/paper/2310.07641/) (2023)
- [MINT: Evaluating LLMs in Multi-turn Interaction with Tools and Language Feedback](https://mlsanity.com/paper/2309.10691/) (2023)
- [LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models](https://mlsanity.com/paper/2310.05736/) (2023)
- [Evaluating the Ripple Effects of Knowledge Editing in Language Models](https://mlsanity.com/paper/2307.12976/) (2023)
- [Continual Learning of Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2404.16789/) (2024)
- [In-context Autoencoder for Context Compression in a Large Language Model](https://mlsanity.com/paper/2307.06945/) (2023)
- [Compressing Context to Enhance Inference Efficiency of Large Language Models](https://mlsanity.com/paper/2310.06201/) (2023)
- [Recall and Learn: Fine-tuning Deep Pretrained Language Models with Less Forgetting](https://mlsanity.com/paper/2004.12651/) (2020)
- [Continual Learning with Pre-Trained Models: A Survey](https://mlsanity.com/paper/2401.16386/) (2024)
- [Benchmarking Complex Instruction-Following with Multiple Constraints Composition](https://mlsanity.com/paper/2407.03978/) (2024)
- [On the Effectiveness of Adapter-based Tuning for Pretrained Language Model Adaptation](https://mlsanity.com/paper/2106.03164/) (2021)
- [DEMix Layers: Disentangling Domains for Modular Language Modeling](https://mlsanity.com/paper/2108.05036/) (2021)
- [FollowBench: A Multi-level Fine-grained Constraints Following Benchmark for Large Language Models](https://mlsanity.com/paper/2310.20410/) (2023)
- [Large Multimodal Agents: A Survey](https://mlsanity.com/paper/2402.15116/) (2024)
- [Multi-IF: Benchmarking LLMs on Multi-Turn and Multilingual Instructions Following](https://mlsanity.com/paper/2410.15553/) (2024)
- [Prompt Compression for Large Language Models: A Survey](https://mlsanity.com/paper/2410.12388/) (2024)
- [Lifelong Language Pretraining with Distribution-Specialized Experts](https://mlsanity.com/paper/2305.12281/) (2023)
- [Learning to Compress Prompt in Natural Language Formats](https://mlsanity.com/paper/2402.18700/) (2024)
- [Injecting Domain-Specific Knowledge into Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2502.10708/) (2025)
- [AGENTIF: Benchmarking Instruction Following of Large Language Models in Agentic Scenarios](https://mlsanity.com/paper/2505.16944/) (2025)

## Built on by (most cited first, 2 of 2)

- [Scaling Laws for Hypernetwork-Based Knowledge Injection in Large Language Models](https://mlsanity.com/paper/2607.19604/) (2026)
- [Following the Flow the Law Requires: Training Voice Agents to Transfer to Unseen Regulated Procedures in Debt Collection](https://mlsanity.com/paper/2610.18881/) (2026)

## Abstract (arXiv)

Large Language Model (LLM)-based agentic systems rely on in-context policy documents encoding diverse business rules. As requirements grow, these documents expand rapidly, causing high computational overhead. This motivates developing internalization methods that embed policy documents into model priors while preserving performance. Prior prompt compression work targets generic prompts, but agentic policy documents span multiple complexity levels and require deeper reasoning, making internalization harder. We introduce CC-Gen, an agentic benchmark generator with Controllable Complexity across four levels, enabling systematic evaluation of agents' ability to handle complexity and offering a unified framework for assessing policy internalization. Our analysis shows that complex policy specifications governing workflows pose major reasoning challenges. Supporting internalization with gold user agent interaction trajectories containing chain-of-thought (CoT) annotations via supervised fine-tuning (SFT) is data-intensive and degrades sharply as policy complexity increases. To mitigate data and reasoning burdens, we propose Category-Aware Policy Continued Pretraining (CAP-CPT). Our automated pipeline parses policy documents to extract key specifications, grouping them into factual, behavioral, and conditional categories, and isolating complex conditions that drive workflow complexity. This guides targeted data synthesis and enables agents to internalize policy information through an autoregressive pretraining loss. Experiments show CAP-CPT improves SFT baselines in all settings, with up to 41% and 22% gains on Qwen-3-32B, achieving 97.3% prompt length reduction on CC-Gen and further enhancing tau-Bench with minimal SFT data.