AI Paper Atlas

Research map / Reinforcement learning / RL from feedback

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

Jiliang Ni, Jiachen Pu, Zhongyi Yang, Jingfeng Luo, Conggang Hu

ICLR 2026 · Reinforcement learning › RL from feedback · approach: RL for LLM reasoning · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Sim-RL refinement after CKD distillation (ablation)CKD student (Qwen3-0.6B) · ACEBench Normalaccuracy: 39.00 -> 53.00Improved
add Sim-RL refinement after CKD distillation (ablation)CKD student (Qwen3-0.6B) · BFCLv3 Overallaccuracy: 49.56 -> 51.70Improved
replace SFT-think initialization with CKD before Sim-RL (ablation)STAR Qwen3-0.6B · BFCLv3 Overallaccuracy: 50.41 -> 51.70Improved
use Sim-RL similarity reward instead of binary reward (ablation)CKD student · ACEBench Normalaccuracy: 35.70 -> 53.00Improved
use Sim-RL similarity reward instead of ToolRL reward (ablation)CKD student · BFCLv3 Overallaccuracy: 48.59 -> 51.70Improved
use Sim-RL similarity reward instead of SWiRL PRM reward (ablation)CKD student · ACEBench Normalaccuracy: 40.30 -> 53.00Improved
use CKD instead of stabilized RKL as distillation before RL (ablation)Qwen3-0.6B · ACEBench Normal (w/ RL)accuracy: 41.30 -> 53.00Improved
use CKD instead of stabilized AKL as distillation before RL (ablation)Qwen3-0.6B · ACEBench Normal (w/ RL)accuracy: 49.00 -> 53.00Improved
use CKD instead of top-k FKL as distillation before RL (ablation)Qwen3-0.6B · ACEBench Normal (w/ RL)accuracy: 50.00 -> 53.00Improved
use SFT instead of no fine-tuningQwen3-0.6B · ACEBench Normal Summaryaccuracy: 27.20 -> 2.10Hurt
increase top-k from 10 to 100 in CKD (ablation)CKD+Sim-RL · ACEBench Normal (w/ RL)accuracy: 49.20 -> 53.00Improved
increase top-k from 100 to 1000 in CKD (ablation)CKD+Sim-RL · ACEBench Normal (w/ RL)accuracy: 53.00 -> 52.20No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy