Beyond Scaling Law: A Data-Efficient Distillation Framework for Reasoning

Large language models (LLMs) demonstrate remarkable reasoning capabilities in tasks such as algorithmic coding and mathematical problem-solving. Recent methods have improved reasoning through expanded corpus and multistage training combining reinforcement learning and supervised fine-tuning. Although some methods suggest that small but targeted dataset can incentivize reasoning via only distillation, a reasoning scaling laws is still taking shape, increasing computational costs. To address this, we propose a data-efficient distillation framework (DED) that optimizes the Pareto frontier of reasoning distillation. Inspired by the on-policy learning and diverse roll-out strategies of reinforcement learning, the key idea of our approach is threefold: (1) We identify that benchmark scores alone do not determine an effective teacher model. Through comprehensive comparisons of leading reasoning LLMs, we develop a method to select an optimal teacher model. (2) While scaling distillation can enhance reasoning, it often degrades out-of-domain performance. A carefully curated, smaller corpus achieves a balanced trade-off between in-domain and out-of-domain capabilities. (3) Diverse reasoning trajectories encourage the student model to develop robust reasoning skills. We validate our method through evaluations on mathematical reasoning (AIME 2024/2025, MATH-500) and code generation (LiveCodeBench), achieving state-of-the-art results with only 0.8k carefully curated examples, bypassing the need for extensive scaling. Our systematic analysis demonstrates that DED outperforms existing methods by considering factors beyond superficial hardness, token length, or teacher model capability. This work offers a practical and efficient pathway to advanced reasoning while preserving general capabilities.

Humanoid Robot-Application and Influence

Application of humanoid robots has been common in the field of healthcare and education. It has been recurrently used to improve social behavior and mollify distress level among children with autism, cancer and cerebral palsy. This article discusses the same from a human factors perspective. It shows how people of different age and gender have a different opinion towards the application and acceptance of humanoid robots. Additionally, this article highlights the influence of cerebral condition and social interaction on a user behavior and attitude towards humanoid robots. Our study performed a literature review and found that (a) children and elderly individuals prefer humanoid robots due to inactive social interaction, (b) The deterministic behavior of humanoid robots can be acknowledged to improve social behavior of autistic children, (c) Trust on humanoid robots is highly driven by its application and a user age, gender, and social life.