SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.00564 · ICLR · 2025

Scaling Offline Model-Based RL via Jointly-Optimized World-Action Model Pretraining

Binhua Li, Yongbin Li, Jie Cheng, Gang Xiong, Qinghai Miao, Yisheng Lv, Ruixi Qiao, Yingwei Ma

arXiv · PDF · Open in the Atlas

Code that ran

10 of 13 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 3 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 7 failures of our instrument; 3 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2410.00564.json · badge: /api/ran/2410.00564.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
CJReinforce/JOWA canonical (classifier label)
classifier basis: census
10 ran · 3 no recorded run
FunctionStatusWhere it lives
batch_tokenize_envs Ran
ran, honoured the contract we drafted
CJReinforce/JOWA/src/action_tokenizer.py
pointer only (licence: GPL-3.0) · get_code("1321dbfc0d5d647b")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

capitalize_game_name Ran
ran; our contract draft was wrong, not the code
CJReinforce/JOWA/results/base_scores.py
pointer only (licence: GPL-3.0) · get_code("03430a4f132483cc")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

collect_scores Ran
ran; our contract draft was wrong, not the code
CJReinforce/JOWA/eval/aggregate_scores.py
pointer only (licence: GPL-3.0) · get_code("1eb71fba6740dd0d")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

grouper Ran
ran; our contract draft was wrong, not the code
CJReinforce/JOWA/src/save_obs_in_token.py
pointer only (licence: GPL-3.0) · get_code("2444d9ba3dc707f5")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

insert_separators Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
CJReinforce/JOWA/src/make_reconstructions.py
pointer only (licence: GPL-3.0) · get_code("289034e68b701092")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

lookup_score_for_env Ran
ran, honoured the contract we drafted
CJReinforce/JOWA/results/base_scores.py
pointer only (licence: GPL-3.0) · get_code("2aac2ff0103e0ddd")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

normalize Ran
ran, honoured the contract we drafted
CJReinforce/JOWA/results/base_scores.py
pointer only (licence: GPL-3.0) · get_code("b4775b8b96581012")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

process_param_name Ran
ran; our contract draft was wrong, not the code
CJReinforce/JOWA/src/save_obs_in_token.py
pointer only (licence: GPL-3.0) · get_code("0b2d9057ed4de466")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

read_last_line Ran
ran; our contract draft was wrong, not the code
CJReinforce/JOWA/eval/aggregate_scores.py
pointer only (licence: GPL-3.0) · get_code("09a6a95dd67b27e7")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

tensor_to_np_frames Ran
ran; our contract draft was wrong, not the code
CJReinforce/JOWA/src/make_reconstructions.py
pointer only (licence: GPL-3.0) · get_code("982a993002d7b356")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

batch_tokenize_actions Not yet run
unverified (harvested, no recorded run)
CJReinforce/JOWA/src/action_tokenizer.py
pointer only (licence: GPL-3.0) · get_code("8dc0dabbb7a9c7fc")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

make_reconstructions_of_trajectories Not yet run
unverified (harvested, no recorded run)
CJReinforce/JOWA/src/make_reconstructions.py
pointer only (licence: GPL-3.0) · get_code("374a8c931eb22f4b")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

tokenize_actions Not yet run
unverified (harvested, no recorded run)
CJReinforce/JOWA/src/action_tokenizer.py
pointer only (licence: GPL-3.0) · get_code("3dec7895772051c7")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, CJReinforce/JOWA, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 12 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
grouper 4 other papers
3 dated earlier, 1 later
1904.05146 (2019-04)
DeepSphere: towards an equivariant graph-based spherical CNN
2106.04799 (2021-06)
Pretraining Representations for Data-Efficient Reinforcement Learning
2406.06037 (2024-06)
Investigating Pre-Training Objectives for Generalization in Vision-Based Reinforcement Learning
and 1 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

A significant aspiration of offline reinforcement learning (RL) is to develop a generalist agent with high capabilities from large and heterogeneous datasets. However, prior approaches that scale offline RL either rely heavily on expert trajectories or struggle to generalize to diverse unseen tasks. Inspired by the excellent generalization of world model in conditional video generation, we explore the potential of image observation-based world model for scaling offline RL and enhancing generalization on novel tasks. In this paper, we introduce JOWA: Jointly-Optimized World-Action model, an offline model-based RL agent pretrained on multiple Atari games with 6 billion tokens data to learn generalpurpose representation and decision-making ability. Our method jointly optimizes a world-action model through a shared transformer backbone, which stabilize temporal difference learning with large models during pretraining. Moreover, we propose a provably efficient and parallelizable planning algorithm to compensate for the Q-value estimation error and thus search out better policies. Experimental results indicate that our largest agent, with 150 million parameters, achieves 78.9% human-level performance on pretrained games using only 10% subsampled offline data, outperforming existing state-of-the-art large-scale offline RL baselines by 71.4% on averange. Furthermore, JOWA scales favorably with model capacity and can sample-efficiently transfer to novel games using only 5k offline fine-tuning data (approximately 4 trajectories) per game, demonstrating superior generalization. The code and checkpoints will be released at https://github.com/CJReinforce/JOWA.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2410.00564")
get_code_for_paper("2410.00564")
have("2410.00564")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2410.00564.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2410.00564.svg)](https://syntology.ai/paper/2410.00564)

Connect an agent — have() is free.