SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.15629 · ICLR · 2024

Language Control Diffusion: Efficiently Scaling through Space, Time, and Tasks

Amy Zhang, William Wang, Yujie Lu, Edwin Zhang, Shinda Huang

arXiv · PDF · Open in the Atlas

Code that ran

2 of 5 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; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 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/2210.15629.json · badge: /api/ran/2210.15629.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
ezhang7423/language-control-diffusion canonical (classifier label)
classifier basis: classifier_2025
2 ran · 3 no recorded run
FunctionStatusWhere it lives
WeightedL1 Ran
ran on a synthesized input
ezhang7423/language-control-diffusion/src/lcd/models/diffusion.py
code served (permissive licence) · get_code("eac173f0a5136904")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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
from torch import nn

class WeightedLoss(nn.Module):

    def __init__(self, weights, action_dim):
        super().__init__()
        self.register_buffer('weights', weights)
        self.action_dim = action_dim

    def forward(self, pred, targ):
        """
        pred, targ : tensor
            [ batch_size x horizon x transition_dim ]
        """
        loss = self._loss(pred, targ)
        weighted_loss = (loss * self.weights).mean()
        a0_loss = (loss[:, 0, :self.action_dim] / self.weights[0, :self.action_dim]).mean()
        return (weighted_loss / self.weights[0, 0].detach(), {'a0_loss': a0_loss})

    def _loss(self, pred, targ):
        raise NotImplementedError

class WeightedL1(WeightedLoss):

    def _loss(self, pred, targ):
        return torch.abs(pred - targ)


batch_size = 2
horizon = 3
transition_dim = 5
action_dim = 2

weights = torch.ones(horizon, transition_dim)
weights[0, :action_dim] = 2.0

INIT_ARGS = {
    'weights': weights,
    'action_dim': action_dim,
}

pred = torch.randn(batch_size, horizon, transition_dim)
targ = torch.randn(batch_size, horizon, transition_dim)

ARGS = {
    'pred': pred,
    'targ': targ,
}

if __name__ == '__main__':
    model = WeightedL1(**INIT_ARGS)
    result = model(**ARGS)
    print("Output:", result)
    print("Weighted loss:", result[0])
    print("A0 loss:", result[1]['a0_loss'])
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
WeightedL2 Ran
ran on a synthesized input
ezhang7423/language-control-diffusion/src/lcd/models/diffusion.py
code served (permissive licence) · get_code("33d5229f504bb6ae")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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
from torch import nn
import torch.nn.functional as F

class WeightedLoss(nn.Module):

    def __init__(self, weights, action_dim):
        super().__init__()
        self.register_buffer('weights', weights)
        self.action_dim = action_dim

    def forward(self, pred, targ):
        """
        pred, targ : tensor
            [ batch_size x horizon x transition_dim ]
        """
        loss = self._loss(pred, targ)
        weighted_loss = (loss * self.weights).mean()
        a0_loss = (loss[:, 0, :self.action_dim] / self.weights[0, :self.action_dim]).mean()
        return (weighted_loss / self.weights[0, 0].detach(), {'a0_loss': a0_loss})

    def _loss(self, pred, targ):
        return NotImplementedError

class WeightedL2(WeightedLoss):

    def _loss(self, pred, targ):
        return F.mse_loss(pred, targ, reduction='none')


batch_size = 2
horizon = 3
transition_dim = 5
action_dim = 2

weights = torch.ones(horizon, transition_dim)
weights[0, :action_dim] = 2.0

INIT_ARGS = {
    'weights': weights,
    'action_dim': action_dim,
}

pred = torch.randn(batch_size, horizon, transition_dim)
targ = torch.randn(batch_size, horizon, transition_dim)

ARGS = {
    'pred': pred,
    'targ': targ,
}

if __name__ == '__main__':
    model = WeightedL2(**INIT_ARGS)
    result = model(**ARGS)
    print("Result:", result)
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
GaussianDiffusion Not yet run
unverified (harvested, no recorded run)
ezhang7423/language-control-diffusion/src/lcd/models/diffusion.py
code served (permissive licence) · get_code("30d9b14f01180374")
How it ran

no invocation recorded.

WeightedLoss Not yet run
unverified (harvested, no recorded run)
ezhang7423/language-control-diffusion/src/lcd/models/diffusion.py
code served (permissive licence) · get_code("d7c1ac55cf6677ba")
How it ran

no invocation recorded.

default_sample_fn Not yet run
unverified (harvested, no recorded run)
ezhang7423/language-control-diffusion/src/lcd/models/diffusion.py
code served (permissive licence) · get_code("64ba75a349d16b60")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 2 later) and 3 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
WeightedL1
brick of this paper's assembly: helper
8 other papers
1 dated earlier, 7 later · 2 only as an assembly brick
2208.06193 (2022-08)
Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning
2305.18459 (2023-05)
Diffusion Model is an Effective Planner and Data Synthesizer for Multi-Task Reinforcement Learning
2309.16984 (2023-09)
Consistency Models as a Rich and Efficient Policy Class for Reinforcement Learning
and 5 more not named here
WeightedL2
brick of this paper's assembly: helper
8 other papers
1 dated earlier, 7 later · 2 only as an assembly brick
2208.06193 (2022-08)
Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning
2305.18459 (2023-05)
Diffusion Model is an Effective Planner and Data Synthesizer for Multi-Task Reinforcement Learning
2309.16984 (2023-09)
Consistency Models as a Rich and Efficient Policy Class for Reinforcement Learning
and 5 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

Training generalist agents is difficult across several axes, requiring us to deal with high-dimensional inputs (space), long horizons (time), and generalization to novel tasks. Recent advances with architectures have allowed for improved scaling along one or two of these axes, but are still computationally prohibitive to use. In this paper, we propose to address all three axes by leveraging Language to Control Diffusion models as a hierarchical planner conditioned on language (LCD). We effectively and efficiently scale diffusion models for planning in extended temporal, state, and task dimensions to tackle long horizon control problems conditioned on natural language instructions, as a step towards generalist agents. Comparing LCD with other state-of-the-art models on the CALVIN language robotics benchmark finds that LCD outperforms other SOTA methods in multi-task success rates, whilst improving inference speed over other comparable diffusion models by 3.3x~15x. We show that LCD can successfully leverage the unique strength of diffusion models to produce coherent long range plans while addressing their weakness in generating low-level details and control.

For agents

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

get_harvested_code_for_paper("2210.15629")
get_code_for_paper("2210.15629")
have("2210.15629")

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

curl https://syntology.ai/api/ran/2210.15629.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.15629.svg)](https://syntology.ai/paper/2210.15629)

Connect an agent — have() is free.