# Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognition

arXiv: https://arxiv.org/abs/2505.22375 · HTML: https://mlsanity.com/paper/2505.22375/

## Key facts

- Cited by 62 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 47 of its 57 references that are in the map.
- First posted on arXiv 28 May 2025.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)

## Builds on (most cited first, 30 of 47)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [A Survey on Large Language Model based Autonomous Agents](https://mlsanity.com/paper/2308.11432/) (2023)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [s1: Simple test-time scaling](https://mlsanity.com/paper/2501.19393/) (2025)
- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [Training Large Language Models to Reason in a Continuous Latent Space](https://mlsanity.com/paper/2412.06769/) (2024)
- [SimpleRL-Zoo: Investigating and Taming Zero Reinforcement Learning for Open Base Models in the Wild](https://mlsanity.com/paper/2503.18892/) (2025)
- [LIMO: Less is More for Reasoning](https://mlsanity.com/paper/2502.03387/) (2025)
- [What Makes Good Data for Alignment? A Comprehensive Study of Automatic Data Selection in Instruction Tuning](https://mlsanity.com/paper/2312.15685/) (2023)
- [From Quantity to Quality: Boosting LLM Performance with Self-Guided Data Selection for Instruction Tuning](https://mlsanity.com/paper/2308.12032/) (2023)
- [Teaching Small Language Models to Reason](https://mlsanity.com/paper/2212.08410/) (2022)
- [A Survey on Knowledge Distillation of Large Language Models](https://mlsanity.com/paper/2402.13116/) (2024)
- [O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning](https://mlsanity.com/paper/2501.12570/) (2025)
- [Are We Done with MMLU?](https://mlsanity.com/paper/2406.04127/) (2024)
- [Distilling System 2 into System 1](https://mlsanity.com/paper/2407.06023/) (2024)
- [Llama-Nemotron: Efficient Reasoning Models](https://mlsanity.com/paper/2505.00949/) (2025)

## Built on by (most cited first, 30 of 62)

- [Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration](https://mlsanity.com/paper/2508.13755/) (2025)
- [Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs](https://mlsanity.com/paper/2507.02076/) (2025)
- [Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey](https://mlsanity.com/paper/2507.09662/) (2025)
- [ROOT: Robust Orthogonalized Optimizer for Neural Network Training](https://mlsanity.com/paper/2511.20626/) (2025)
- [Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking](https://mlsanity.com/paper/2509.23392/) (2025)
- [VAR-MATH: Probing True Mathematical Reasoning in LLMS via Symbolic Multi-Instance Benchmarks](https://mlsanity.com/paper/2507.12885/) (2025)
- [CAMERA: Multi-Matrix Joint Compression for MoE Models via Micro-Expert Redundancy Analysis](https://mlsanity.com/paper/2508.02322/) (2025)
- [What Matters For Safety Alignment?](https://mlsanity.com/paper/2601.03868/) (2026)
- [Streaming-dLLM: Accelerating Diffusion LLMs via Suffix Pruning and Dynamic Decoding](https://mlsanity.com/paper/2601.17917/) (2026)
- [Fine-tuning Large Language Model for Automated Algorithm Design](https://mlsanity.com/paper/2507.10614/) (2025)
- [Revealing the Power of Post-Training for Small Language Models via Knowledge Distillation](https://mlsanity.com/paper/2509.26497/) (2025)
- [Efficient Reinforcement Learning with Semantic and Token Entropy for LLM Reasoning](https://mlsanity.com/paper/2512.04359/) (2025)
- [Ratio-Variance Regularized Policy Optimization for Efficient LLM Fine-tuning](https://mlsanity.com/paper/2601.03320/) (2026)
- [DLLM Agent: See Farther, Run Faster](https://mlsanity.com/paper/2602.07451/) (2026)
- [Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression](https://mlsanity.com/paper/2602.08324/) (2026)
- [From Prefix Cache to Fusion RAG Cache: Accelerating LLM Inference in Retrieval-Augmented Generation](https://mlsanity.com/paper/2601.12904/) (2026)
- [Enhancing the Medical Context-Awareness Ability of LLMs via Multifaceted Self-Refinement Learning](https://mlsanity.com/paper/2511.10067/) (2025)
- [AGZO: Activation-Guided Zeroth-Order Optimization for LLM Fine-Tuning](https://mlsanity.com/paper/2601.17261/) (2026)
- [Taxonomy-Aware Representation Alignment for Hierarchical Visual Recognition with Large Multimodal Models](https://mlsanity.com/paper/2603.00431/) (2026)
- [Efficient Reasoning with Balanced Thinking](https://mlsanity.com/paper/2603.12372/) (2026)
- [OPERA: An Agent for Image Restoration with End-to-End Joint Planning-Execution Optimization](https://mlsanity.com/paper/2605.22104/) (2026)
- [Searching Meta Reasoning Skeleton to Guide LLM Reasoning](https://mlsanity.com/paper/2510.04116/) (2025)
- [W4A16 Mixed-Precision Matrix Multiplication on Decoupled Architecture: Kernel Design and Memory Bottleneck Analysis for Ascend NPUs](https://mlsanity.com/paper/2601.16536/) (2026)
- [Near-Policy: Accelerating On-Policy Distillation via Asynchronous Generation and Selective Packing](https://mlsanity.com/paper/2605.05940/) (2026)
- [MoRAgent: Parameter Efficient Agent Tuning with Mixture-of-Roles](https://mlsanity.com/paper/2512.21708/) (2025)
- [TiFRe: Text-guided Video Frame Reduction for Efficient Video Multi-modal Large Language Models](https://mlsanity.com/paper/2602.08861/) (2026)
- [Differentially Private and Communication Efficient Large Language Model Split Inference via Stochastic Quantization and Soft Prompt](https://mlsanity.com/paper/2602.11513/) (2026)
- [SpatialText: A Pure-Text Cognitive Benchmark for Spatial Understanding in Large Language Models](https://mlsanity.com/paper/2603.03002/) (2026)
- [Fairness-Aware and Latency-Controllable Scheduling for Chunked-Prefill LLM Serving](https://mlsanity.com/paper/2606.09061/) (2026)
- [Post-Training Quantization of OpenPangu Models for Efficient Deployment on Atlas A2](https://mlsanity.com/paper/2512.23367/) (2025)

## Abstract (arXiv)

This work presents Pangu Embedded, an efficient Large Language Model (LLM) reasoner developed on Ascend Neural Processing Units (NPUs), featuring flexible fast and slow thinking capabilities. Pangu Embedded addresses the significant computational costs and inference latency challenges prevalent in existing reasoning-optimized LLMs. We propose a two-stage training framework for its construction. In Stage 1, the model is finetuned via an iterative distillation process, incorporating inter-iteration model merging to effectively aggregate complementary knowledge. This is followed by reinforcement learning on Ascend clusters, optimized by a latency-tolerant scheduler that combines stale synchronous parallelism with prioritized data queues. The RL process is guided by a Multi-source Adaptive Reward System (MARS), which generates dynamic, task-specific reward signals using deterministic metrics and lightweight LLM evaluators for mathematics, coding, and general problem-solving tasks. Stage 2 introduces a dual-system framework, endowing Pangu Embedded with a "fast" mode for routine queries and a deeper "slow" mode for complex inference. This framework offers both manual mode switching for user control and an automatic, complexity-aware mode selection mechanism that dynamically allocates computational resources to balance latency and reasoning depth. Experimental results on benchmarks including AIME 2024, GPQA, and LiveCodeBench demonstrate that Pangu Embedded with 7B parameters, outperforms similar-size models like Qwen3-8B and GLM4-9B. It delivers rapid responses and state-of-the-art reasoning quality within a single, unified model architecture, highlighting a promising direction for developing powerful yet practically deployable LLM reasoners.