In Their Own Words: Reasoning Traces Tailored for Small Models Make Them Better Reasoners

arXiv:2509.22230 · cs.CL · Submitted 2025-09-26 · Read on arXiv

cs.CL

Submitted: 2025-09-26

Updated: 2026-09-26

Code: https://github.com/jaeh8nkim/equigranular

Terminology

• Trace — 128× in this paper · explained in 2 episodes
• Alignment — 25× in this paper · explained in 1 episode
• Reasoning Traces — 24× in this paper · explained in 3 episodes
• Distillation — 13× in this paper · explained in 2 episodes
• Benchmark — 12× in this paper · explained in 2 episodes
• Speculative Decoding — 10× in this paper · explained in 2 episodes
• Perplexity — 7× in this paper · explained in 3 episodes
• Reverse Speculative Decoding — 7× in this paper
• Accuracy — 6× in this paper · explained in 1 episode
• Cognitive Load — 6× in this paper · explained in 2 episodes
• Fine-Tuning — 5× in this paper · explained in 5 episodes
• Self-distillation — 5× in this paper · explained in 6 episodes
• Reinforcement Learning — 3× in this paper · explained in 4 episodes
• Cross-Model Transfer — 2× in this paper · explained in 2 episodes
• Knowledge Distillation — 2× in this paper · explained in 11 episodes
• Adam — 1× in this paper · explained in 2 episodes
• Core — 1× in this paper · explained in 2 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 2 episodes
• Inspired by the Unsupervised Prefix Fine-Tuning — 1× in this paper
• Pruning — 1× in this paper · explained in 4 episodes
• Reinforcement Learning (RL) — 1× in this paper · explained in 25 episodes
• Speculative Knowledge Distillation — 1× in this paper
• Structured Reasoning — 1× in this paper · explained in 3 episodes
• Supervised Fine-Tuning (SFT) — 1× in this paper · explained in 8 episodes
• Transformer — 1× in this paper · explained in 2 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs
• SplitReason: Learning To Offload Reasoning
• Role of obstacle softness in the diffusive behavior of active Particles
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• Gemma 3 Technical Report
• The First Few Tokens Are All You Need: An Efficient and Effective Unsupervised Prefix Fine-Tuning Method for Reasoning Models
• Stationary flows for viscous heat-conductive fluid in a perturbed half-space
• Stabilizing Knowledge, Promoting Reasoning: Dual-Token Constraints for RLVR
• Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning
• Effect of Dielectric Wakefields in a Capillary Discharge for Plasma Wakefield Acceleration
• Phi-4-Mini-Reasoning: Exploring the Limits of Small Reasoning Language Models in Math
• Qwen2.5 Technical Report
• Speculative Thinking: Enhancing Small-Model Reasoning with Large Model Guidance at Inference Time
• Scaling Relationship on Learning Mathematical Reasoning with Large Language Models
• Transformers Utilization in Chart Understanding: A Review of Recent Advances & Future Trends
• Short-time Fourier Transform-based Signal Recovery for Modulo Analog-to-Digital Converters

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering