SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.18043 · NeurIPS · 2024

GenRL: Multimodal-foundation world models for generalization in embodied agents

Tim Verbelen, Bart Dhoedt, Aaron Courville, Pietro Mazzaglia, Sai Rajeswar

arXiv · PDF · Open in the Atlas

Code that ran

6 of 9 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 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/2406.18043.json · badge: /api/ran/2406.18043.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
mazpie/genrl canonical (classifier label)
classifier basis: census
6 ran · 3 no recorded run
FunctionStatusWhere it lives
env_reward Ran
ran; our contract draft was wrong, not the code
mazpie/genrl/agent/dreamer.py
code served (permissive licence) · get_code("29ed0f8937fe254f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned builtin_function_or_method. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

class RewardHead(nn.Module):
    def __init__(self, feat_dim, output_dim=1):
        super().__init__()
        self.linear = nn.Linear(feat_dim, output_dim)
    
    def forward(self, x):
        return self.linear(x)

class WorldModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.heads = nn.ModuleDict({
            'reward': RewardHead(feat_dim=4, output_dim=1)
        })

class Agent(nn.Module):
    def __init__(self):
        super().__init__()
        self.wm = WorldModel()

agent = Agent()

seq = {
    'feat': torch.randn(2, 4)
}

ARGS = {
    'agent': agent,
    'seq': seq
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type builtin_function_or_method — a builtin_function_or_method output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a builtin_function_or_method output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
make_dreamer_agent Ran
ran; our contract draft was wrong, not the code
mazpie/genrl/collect_data.py
code served (permissive licence) · get_code("55a2143c42c8051e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from dataclasses import dataclass, field
from typing import Dict, Any
from omegaconf import DictConfig, OmegaConf

torch.manual_seed(0)

# Create a minimal obs_space
@dataclass
class ObsSpace:
    shape: tuple = (4,)
    dtype: type = float

obs_space = ObsSpace()

# Create a minimal action_spec
@dataclass
class ActionSpec:
    shape: tuple = (3,)
    dtype: type = float
    minimum: float = -1.0
    maximum: float = 1.0

action_spec = ActionSpec()

# Create cur_config with an agent field that will be deleted
cur_config_dict = {
    "agent": {
        "type": "dreamer",
        "hidden_size": 64
    },
    "model": {
        "layers": 2,
        "units": 32
    },
    "training": {
        "epochs": 10,
        "batch_size": 2
    }
}
cur_config = OmegaConf.create(cur_config_dict)

# Create cfg that will be passed to hydra.utils.instantiate
# This should be a DictConfig with _target_ for instantiation
cfg_dict = {
    "_target_": "builtins.dict",
    "model_size": 128,
    "learning_rate": 0.001
}
cfg = OmegaConf.create(cfg_dict)

ARGS = {
    "obs_space": obs_space,
    "action_spec": action_spec,
    "cur_config": cur_config,
    "cfg": cfg
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
hydra, omegaconf, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
signed_hyperbolic Ran
ran; our contract draft was wrong, not the code
mazpie/genrl/agent/dreamer_utils.py
code served (permissive licence) · get_code("a4ac0b3ee7c693f9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create a small tensor for the input
x = torch.randn(2, 4)

# Define the eps parameter
eps = 0.001

# Construct the ARGS dictionary
ARGS = {
    "x": x,
    "eps": eps
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
stop_gradient Ran
ran, honoured the contract we drafted
mazpie/genrl/agent/dreamer.py
code served (permissive licence) · get_code("17f4154f5359d704")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function stop_gradient takes a tensor x and calls x.detach() on it
# x must be a torch.Tensor

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
symexp Ran
ran; our contract draft was wrong, not the code
mazpie/genrl/agent/dreamer_utils.py
code served (permissive licence) · get_code("fd421307a7e4cede")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function symexp takes a single parameter x and applies:
# torch.sign(x) * (torch.exp(torch.abs(x)) - 1.0)
# So x should be a tensor

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
symlog Ran
ran; our contract draft was wrong, not the code
mazpie/genrl/agent/dreamer_utils.py
code served (permissive licence) · get_code("3e346b9b00ea8169")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function symlog takes a single parameter x and applies:
# torch.sign(x), torch.log(torch.abs(x) + 1.0)
# So x should be a tensor that supports these operations

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
connector_update_fn Not yet run
unverified (harvested, no recorded run)
mazpie/genrl/agent/genrl.py
code served (permissive licence) · get_code("30b4702622c6564f")
How it ran

no invocation recorded.

make Not yet run
unverified (harvested, no recorded run)
mazpie/genrl/envs/custom_dmc_tasks/cheetah.py
code served (permissive licence) · get_code("bf18772e2ed3e413")
How it ran

no invocation recorded.

make_agent Not yet run
unverified (harvested, no recorded run)
mazpie/genrl/collect_data.py
code served (permissive licence) · get_code("2a76dfe317adeb03")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 1 the same month, 3 later) and 4 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
make_agent 9 other papers
7 dated earlier, 1 the same month, 1 later
2103.04551 (2021-03)
Behavior From the Void: Unsupervised Active Pre-Training
2110.15191 (2021-10)
URLB: Unsupervised Reinforcement Learning Benchmark
2202.00161 (2022-02)
CIC: Contrastive Intrinsic Control for Unsupervised Skill Discovery
and 6 more not named here
symexp 5 other papers
4 dated earlier, 1 later
2210.13702 (2022-10)
DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality
2301.04104 (2023-01)
Mastering Diverse Domains through World Models
2310.05167 (2023-10)
Hieros: Hierarchical Imagination on Structured State Space Sequence World Models
and 2 more not named here
symlog 5 other papers
4 dated earlier, 1 later
2210.13702 (2022-10)
DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality
2301.04104 (2023-01)
Mastering Diverse Domains through World Models
2310.05167 (2023-10)
Hieros: Hierarchical Imagination on Structured State Space Sequence World Models
and 2 more not named here
stop_gradient 2 other papers
2 dated earlier
2209.12016 (2022-09)
Mastering the Unsupervised Reinforcement Learning Benchmark from Pixels
2402.07871 (2024-02)
Scaling Laws for Fine-Grained Mixture of Experts
make_dreamer_agent 1 other paper
1 dated earlier
2405.14073 (2024-05)
PEAC: Unsupervised Pre-training for Cross-Embodiment Reinforcement Learning

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

Multimodal-foundation world models connect and align the video-language space of a foundation model with the latent space of a generative world model for reinforcement learning, requiring vision-only data. Our GenRL framework turns visual and/or language prompts into latent targets and learns to realize the corresponding behaviors by training in the world model's imagination.

For agents

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

get_harvested_code_for_paper("2406.18043")
get_code_for_paper("2406.18043")
have("2406.18043")

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

curl https://syntology.ai/api/ran/2406.18043.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.18043.svg)](https://syntology.ai/paper/2406.18043)

Connect an agent — have() is free.