STAR: Similarity-guided Teacher-Assisted Refinement for Super-Tiny Function Calling Models

• URL: http://arxiv.org/abs/2602.03022v1
• Date: Tue, 03 Feb 2026 02:41:11 GMT
• Title: STAR: Similarity-guided Teacher-Assisted Refinement for Super-Tiny Function Calling Models
• Authors: Jiliang Ni, Jiachen Pu, Zhongyi Yang, Jingfeng Luo, Conggang Hu,
• Abstract summary: Large Language Models (LLMs) in function calling are pivotal for creating advanced AI agents, yet their large scale hinders widespread adoption.<n>We introduce STAR: Similarity-guided Teacher-Assisted Refinement, a novel holistic framework that effectively transfers LLMs' capabilities to super-tiny models.
• Score: 0.7829352305480285
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The proliferation of Large Language Models (LLMs) in function calling is pivotal for creating advanced AI agents, yet their large scale hinders widespread adoption, necessitating transferring their capabilities into smaller ones. However, existing paradigms are often plagued by overfitting, training instability, ineffective binary rewards for multi-solution tasks, and the difficulty of synergizing techniques. We introduce STAR: Similarity-guided Teacher-Assisted Refinement, a novel holistic framework that effectively transfers LLMs' capabilities to super-tiny models. STAR consists of two core technical innovations: (1) Constrained Knowledge Distillation (CKD), a training objective that augments top-k forward KL divergence to suppress confidently incorrect predictions, ensuring training stability while preserving exploration capacity for downstream RL. STAR holistically synergizes these strategies within a cohesive training curriculum, enabling super-tiny models to achieve exceptional performance on complex function calling tasks; (2) Similarity-guided RL (Sim-RL), a RL mechanism that introduces a fine-grained, similarity-based reward. This provides a robust, continuous, and rich signal for better policy optimization by evaluating the similarity between generated outputs and the ground truth. Extensive experiments on challenging and renowned benchmarks demonstrate the effectiveness of our method. Our STAR models establish SOTA in their size classes, significantly outperforming baselines. Remarkably, our 0.6B STAR model achieves the best performance among all open models under 1B, surpassing even several well-known open models at a larger scale. STAR demonstrates a training framework that distills capabilities of LLMs into super-tiny models, paving the way for powerful, accessible, and efficient AI agents.

Related papers

• Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Models [78.73992315826035]
  We introduce Youtu-LLM, a lightweight language model that harmonizes high computational efficiency with native agentic intelligence.<n>Youtu-LLM is pre-trained from scratch to systematically cultivate reasoning and planning capabilities.
  arXiv  Detail & Related papers  (2025-12-31T04:25:11Z)
• STAR: STacked AutoRegressive Scheme for Unified Multimodal Learning [37.68078190711403]
  We introduce STAR: a STacked AutoRegressive scheme for task-progressive unified multimodal learning.<n>This approach decomposes multimodal learning into multiple stages: understanding, generation, and editing.<n> Experiments demonstrate that STAR achieves state-of-the-art performance on GenEval (0.91), DPG-Bench (87.44), and ImgEdit (4.34)
  arXiv  Detail & Related papers  (2025-12-15T07:02:59Z)
• MENTOR: A Reinforcement Learning Framework for Enabling Tool Use in Small Models via Teacher-Optimized Rewards [8.645370827540996]
  Distilling the tool-using capabilities of large language models (LLMs) into smaller, more efficient small language models (SLMs) is a key challenge for their practical application.<n>The predominant approach, supervised fine-tuning (SFT), suffers from poor generalization as it trains models to imitate a static set of teacher trajectories rather than learn a robust methodology.<n>We propose MENTOR, a framework that combines reinforcement learning with teacher-guided distillation.
  arXiv  Detail & Related papers  (2025-10-21T08:03:14Z)
• Revealing the Power of Post-Training for Small Language Models via Knowledge Distillation [43.68215777330875]
  We introduce a systematic post-training pipeline that efficiently enhances small model accuracy.<n>The resulting instruction-tuned model achieves state-of-the-art performance.<n>This work provides a practical and efficient solution for developing high-performance language models on Ascend edge devices.
  arXiv  Detail & Related papers  (2025-09-30T16:40:55Z)
• Metis-RISE: RL Incentivizes and SFT Enhances Multimodal Reasoning Model Learning [20.515599491717442]
  We introduce textbfMetis-RISE (textbfRL textbfSFT textbfEnhances) for multimodal reasoning model learning.
  arXiv  Detail & Related papers  (2025-06-16T02:56:13Z)
• Diffusion Guidance Is a Controllable Policy Improvement Operator [98.11511661904618]
  CFGRL is trained with the simplicity of supervised learning, yet can further improve on the policies in the data.<n>On offline RL tasks, we observe a reliable trend -- increased guidance weighting leads to increased performance.
  arXiv  Detail & Related papers  (2025-05-29T14:06:50Z)
• Multiple Weaks Win Single Strong: Large Language Models Ensemble Weak Reinforcement Learning Agents into a Supreme One [28.264011412168347]
  Model ensemble is a useful approach in reinforcement learning (RL) for training effective agents.<n>We propose LLM-Ens, a novel approach that enhances RL model ensemble with task-specific semantic understandings.
  arXiv  Detail & Related papers  (2025-05-21T09:35:43Z)
• On the Diminishing Returns of Complex Robust RAG Training in the Era of Powerful LLMs [85.688901949146]
  We investigate the question: does the benefit of complex robust training methods diminish as language models become more powerful?<n>Our analysis reveals a consistent trend: emphthe marginal robustness benefit of sophisticated training strategies decreases substantially as model capacity increases.<n>Further investigation demonstrates that stronger models naturally exhibit better confidence calibration, cross-dataset generalization capability, and more effective attention patterns, even under simple training regimes.
  arXiv  Detail & Related papers  (2025-02-17T03:34:31Z)
• Vintix: Action Model via In-Context Reinforcement Learning [78.11130335098936]
  We present the first steps toward scaling ICRL by introducing a fixed, cross-domain model capable of learning behaviors through in-context reinforcement learning.<n>Our results demonstrate that Algorithm Distillation, a framework designed to facilitate ICRL, offers a compelling and competitive alternative to expert distillation to construct versatile action models.
  arXiv  Detail & Related papers  (2025-01-31T18:57:08Z)
• SALMON: Self-Alignment with Instructable Reward Models [80.83323636730341]
  This paper presents a novel approach, namely SALMON, to align base language models with minimal human supervision. We develop an AI assistant named Dromedary-2 with only 6 exemplars for in-context learning and 31 human-defined principles.
  arXiv  Detail & Related papers  (2023-10-09T17:56:53Z)
• One-stop Training of Multiple Capacity Models [74.87789190840527]
  We propose a novel one-stop training framework to jointly train high-capacity and low-capactiy models. Unlike knowledge distillation, where multiple capacity models are trained from scratch separately, our approach integrates supervisions from different capacity models simultaneously.
  arXiv  Detail & Related papers  (2023-05-23T13:44:09Z)

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.