Phi-4-Mini-Reasoning: Exploring the Limits of Small Reasoning Language Models in Math

• URL: http://arxiv.org/abs/2504.21233v1
• Date: Wed, 30 Apr 2025 00:04:35 GMT
• Title: Phi-4-Mini-Reasoning: Exploring the Limits of Small Reasoning Language Models in Math
• Authors: Haoran Xu, Baolin Peng, Hany Awadalla, Dongdong Chen, Yen-Chun Chen, Mei Gao, Young Jin Kim, Yunsheng Li, Liliang Ren, Yelong Shen, Shuohang Wang, Weijian Xu, Jianfeng Gao, Weizhu Chen,
• Abstract summary: Chain-of-Thought (CoT) significantly enhances formal reasoning capabilities in Large Language Models (LLMs)<n>However, improving reasoning in Small Language Models (SLMs) remains challenging due to their limited model capacity.<n>We present a systematic training recipe for SLMs that consists of four steps: (1) large-scale mid-training on diverse distilled long-CoT data, (2) supervised fine-tuning on high-quality long-CoT data, (3) Rollout DPO leveraging a carefully curated preference dataset, and (4) Reinforcement Learning (RL) with Verifiable Reward.
• Score: 135.1260782461186
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Chain-of-Thought (CoT) significantly enhances formal reasoning capabilities in Large Language Models (LLMs) by training them to explicitly generate intermediate reasoning steps. While LLMs readily benefit from such techniques, improving reasoning in Small Language Models (SLMs) remains challenging due to their limited model capacity. Recent work by Deepseek-R1 demonstrates that distillation from LLM-generated synthetic data can substantially improve the reasoning ability of SLM. However, the detailed modeling recipe is not disclosed. In this work, we present a systematic training recipe for SLMs that consists of four steps: (1) large-scale mid-training on diverse distilled long-CoT data, (2) supervised fine-tuning on high-quality long-CoT data, (3) Rollout DPO leveraging a carefully curated preference dataset, and (4) Reinforcement Learning (RL) with Verifiable Reward. We apply our method on Phi-4-Mini, a compact 3.8B-parameter model. The resulting Phi-4-Mini-Reasoning model exceeds, on math reasoning tasks, much larger reasoning models, e.g., outperforming DeepSeek-R1-Distill-Qwen-7B by 3.2 points and DeepSeek-R1-Distill-Llama-8B by 7.7 points on Math-500. Our results validate that a carefully designed training recipe, with large-scale high-quality CoT data, is effective to unlock strong reasoning capabilities even in resource-constrained small models.

Related papers

• Phi-4-reasoning Technical Report [42.508165017775]
  We introduce Phi-4-reasoning, a 14-billion parameter reasoning model that achieves strong performance on complex reasoning tasks.<n>We develop Phi-4-reasoning-plus, a variant enhanced through a short phase of outcome-based reinforcement learning.<n>Both models outperform significantly larger open-weight models such as DeepSeek-R1-Distill-Llama-70B model and approach the performance levels of full DeepSeek-R1 model.
  arXiv  Detail & Related papers  (2025-04-30T05:05:09Z)
• ToM-RL: Reinforcement Learning Unlocks Theory of Mind in Small LLMs [14.29992535286614]
  We show that rule-based reinforcement learning can unlock Theory of Mind (ToM) reasoning capabilities even in small-scale language models.<n>Our RL-trained 7B model achieves 84.50% accuracy on the Hi-ToM benchmark, surpassing models like GPT-4o and DeepSeek-v3.<n>These findings highlight RL's potential to enhance social cognitive reasoning, bridging the gap between structured problem-solving and nuanced social inference.
  arXiv  Detail & Related papers  (2025-04-02T12:58:42Z)
• Reinforcement Learning for Reasoning in Small LLMs: What Works and What Doesn't [0.0]
  Our study investigates the potential of reinforcement learning to improve reasoning in small language models (LLMs)<n>Training on 4 NVIDIA A40 GPUs (48 GB VRAM each) within 24 hours resulted in rapid reasoning gains.<n>These findings highlight the efficacy of RL-based fine-tuning for small LLMs, offering a cost-effective alternative to large-scale approaches.
  arXiv  Detail & Related papers  (2025-03-20T15:13:23Z)
• Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs [195.24565517943802]
  We introduce Phi-4-Mini and Phi-4-Multimodal, compact yet highly capable language and multimodal models. Phi-4-Mini is a 3.8-billion- parameter language model trained on high-quality web and synthetic data. Phi-4-Multimodal is a multimodal model that integrates text, vision, and speech/audio input modalities into a single model.
  arXiv  Detail & Related papers  (2025-03-03T17:05:52Z)
• Unveiling the Key Factors for Distilling Chain-of-Thought Reasoning [33.02060729778806]
  This study examines the factors influencing Chain-of-Thought (CoT) distillation in Small Language Models (SLMs) We find that SLMs exhibit a non-monotonic relationship with granularity, with stronger models benefiting from finer-grained reasoning and weaker models performing better with simpler CoT supervision. These findings emphasize the need to tailor CoT strategies to specific student model, offering actionable insights for optimizing CoT distillation in SLMs.
  arXiv  Detail & Related papers  (2025-02-25T09:08:45Z)
• MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies [85.57899012821211]
  Small Language Models (SLMs) are a resource-efficient alternative to Large Language Models (LLMs) We introduce MiniCPM, specifically the 1.2B and 2.4B non-embedding parameter variants. We also introduce MiniCPM family, including MiniCPM-DPO, MiniCPM-MoE and MiniCPM-128K.
  arXiv  Detail & Related papers  (2024-04-09T15:36:50Z)
• Mixed Distillation Helps Smaller Language Model Better Reasoning [27.934081882868902]
  We introduce Mixed Distillation (MD) framework, which capitalizes on the strengths of Program of Thought (PoT) and Chain of Thought (CoT) capabilities within large language models (LLMs) Our experimental results show that MD significantly enhances the single-path and multi-path reasoning ability of smaller models in various tasks.
  arXiv  Detail & Related papers  (2023-12-17T14:28:28Z)
• Retrieval-based Knowledge Transfer: An Effective Approach for Extreme Large Language Model Compression [64.07696663255155]
  Large-scale pre-trained language models (LLMs) have demonstrated exceptional performance in various natural language processing (NLP) tasks. However, the massive size of these models poses huge challenges for their deployment in real-world applications. We introduce a novel compression paradigm called Retrieval-based Knowledge Transfer (RetriKT) which effectively transfers the knowledge of LLMs to extremely small-scale models.
  arXiv  Detail & Related papers  (2023-10-24T07:58:20Z)
• Knowledge-Augmented Reasoning Distillation for Small Language Models in Knowledge-Intensive Tasks [90.11273439036455]
  Large Language Models (LLMs) have shown promising performance in knowledge-intensive reasoning tasks. We propose Knowledge-Augmented Reasoning Distillation (KARD), a novel method that fine-tunes small LMs to generate rationales from LLMs with augmented knowledge retrieved from an external knowledge base. We empirically show that KARD significantly improves the performance of small T5 and GPT models on the challenging knowledge-intensive reasoning datasets.
  arXiv  Detail & Related papers  (2023-05-28T13:00:00Z)
• PaD: Program-aided Distillation Can Teach Small Models Reasoning Better than Chain-of-thought Fine-tuning [20.59775450213501]
  We propose Program-aided Distillation (PaD), which introduces reasoning programs to suppress the errors in distilled data. We evaluate PaD on arithmetic reasoning, symbolic reasoning, and general ability.
  arXiv  Detail & Related papers  (2023-05-23T10:11:56Z)
• Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes [91.58845026796149]
  We introduce Distilling step-by-step, a new mechanism that trains small models that outperform large language models. We present three findings across 4 NLP benchmarks.
  arXiv  Detail & Related papers  (2023-05-03T17:50:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.