AI Paper Atlas

Research map / Large language models / Agents, code and math

On Training Large Language Models for Long-Horizon Tasks: An Empirical Study of Horizon Length

Sunghwan Kim, Junhee Cho, Beong-woo Kwak, Taeyoon Kwon, Liang Wang, Nan Yang, Xingxing Zhang, Furu Wei, Jinyoung Yeo

ICML 2026 · Large language models › Agents, code and math · approach: Long-horizon LLM agents · 0 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
use macro actions (multiple atomic actions per step) instead of atomic actionsQwen3-1.7B SFT+RL agent · Sudoku L3-L4 (long goal distance)average test success rateImproved
use macro actions instead of atomic actionsQwen3-1.7B SFT+RL agent · Sudoku and Rush Hour L1-L2convergence speed and final success rateImproved
restrict macro-action policy to one atomic step per turn (restore long horizon) (ablation)macro-action policy · Sudokutraining stabilityHurt
use flexible-length macro actions rather than fixed-length (n=5) macros (ablation)GPT-5-mini, Gemini-3-Flash on Sudoku · Sudokusuccess rateImproved
add subgoal decomposition with dense subgrid rewards and per-segment returnsRL on Sudoku · Sudoku L3-L4average success rateImproved
horizon reduction on WebShopQwen3-1.7B RL agent · WebShoptraining stability and average success rateImproved
scale model from 1.7B to 4B with atomic actions (ablation)RL on Sudoku L3-L4 · Sudoku L3-L4training stabilityNo clear effect
add horizon reduction to a 4B modelQwen3 4B RL on Sudoku · Sudoku L3-L4collapse and performanceImproved
use GRPO-style optimizer instead of REINFORCE variant, with and without horizon reduction (ablation)RL on Sudoku · Sudokusuccess rate over trainingImproved
train on L1-L4 range with macro actions and test on longer unseen goal distancesmacro-action policy · Sudoku and Rush Hour unseen horizonssuccess rateImproved
add curriculum: train on short distances (4-9) then continue on long (10-12) (ablation)RL on Rush Hour · Rush Hoursuccess rateImproved

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