Lune

ACL Findings2026

TRM-Planner: Offline Target Planning and Distillation for Tiny Recursive Models

Euijin Baek, Housam Khalifa Bashier Babiker, Mi-Young Kim, Randy Goebel

Abstract

Tiny Recursive Models (TRMs) perform iterative reasoning with an Adaptive Computation Time (ACT)-style loop, but their effective supervision can be target-sensitive to refinement depth and halting calibration, and their halting behavior can be difficult to tune. We introduce TRM-Planner, a two-stage teachercache distillation recipe that shifts compute to an offline teacher-cache stage. A frozen TRM checkpoint is unrolled for multiple refinement steps and stochastic rollouts; for each instance we cache a small set of teacher entries (tokens, logits, step index, and quality metadata). A student TRM is then trained with the standard TRM objective plus a distillation loss computed from cached entries. Across Sudoku-Extreme and ARC-AGI-1/2, TRM-Planner shows an improvement over our reproduced TRM baseline while leaving studenttime inference unchanged. On ARC1/ARC2 with 7M parameters, the two-attempt accuracy (pass@2) increases from 43.1%→48.1% and 6.7%→9.2%, respectively.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

Builds onCited byLeft to right by year; the more cited, the higher.

Builds on2

  1. Grokking at the Edge of Numerical StabilityLucas Prieto, Melih Barsbey, Pedro A. M. Mediano, Tolga BirdalICLR 2025
  2. Self-Training With Noisy Student Improves ImageNet ClassificationQizhe Xie, Minh-Thang Luong, Eduard H. Hovy, Quoc V. LeCVPR 2020

Related papers

  1. Symbol-Equivariant Recurrent Reasoning ModelsRichard Freinschlag, Timo Bertram, Erich Kobler, Andreas Mayr et al.ICML 2026
  2. Enhancing Language Model Reasoning with Structured Multi-Level ModelingSiheng Xiong, Ali Payani, Faramarz FekriICLR 2026
  3. One Step Forward and K Steps Back: Better Reasoning with Denoising Recursion ModelsChris Cameron, Wangzheng Wang, Nikita Ivanov, Ashmita Bhattacharyya et al.ICML 2026
  4. Teach Small Models to Reason by Curriculum DistillationWangyi Jiang, Yaojie Lu, Hongyu Lin, Xianpei Han et al.EMNLP 2025
  5. DRP: Distilled Reasoning Pruning with Mathematical Skill-aware Step Decomposition for Efficient Large Reasoning ModelsYuxuan Jiang, Dawei Li, Francis FerraroACL Findings 2026 · 2 citations
Dusk over the sea between two cliffs drawn in fine vertical lines