SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2103.07854 · ICCV · 2021

Three Steps to Multimodal Trajectory Prediction: Modality Clustering, Classification and Synthesis

Cewu Lu, Jianhua Sun, Yuxuan Li, Hao-Shu Fang

arXiv · PDF · Open in the Atlas

Code that ran

5 of 6 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 5 have no contract verdict; 0 failures of our instrument; 1 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/2103.07854.json · badge: /api/ran/2103.07854.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
ApeironY/PCCSNet community
Papers with Code archive
5 ran · 1 no recorded run
FunctionStatusWhere it lives
Classifier Ran
ran on a synthesized input
ApeironY/PCCSNet/PCCSNet/models/components.py
code served (permissive licence) · get_code("4b986c6b160f6fb6")
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 Tensor, ndim 2, dtype family float, shape (2, 3) 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
import torch.nn as nn

class Classifier(nn.Module):

    def __init__(self, in_features, n_clusters, mid_size=128):
        super(Classifier, self).__init__()
        self.classifier = nn.Sequential(nn.Linear(in_features, mid_size), nn.Tanh(), nn.Linear(mid_size, mid_size), nn.Tanh())
        self.top_layer = nn.Linear(mid_size, n_clusters)

    def forward(self, data):
        data = self.classifier(data)
        data = self.top_layer(data)
        return torch.softmax(data, dim=1)


INIT_ARGS = {
    'in_features': 4,
    'n_clusters': 3,
    'mid_size': 128
}

ARGS = {
    'data': torch.randn(2, 4)
}

if __name__ == '__main__':
    model = Classifier(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Output:\n{output}")
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 Tensor, ndim 2, dtype family float, shape (2, 3) — shape (2, 3) 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
DecoderLSTM Ran
ran on a synthesized input
ApeironY/PCCSNet/PCCSNet/models/components.py
code served (permissive licence) · get_code("1a81b03ff9b7e0f6")
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. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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

# INIT_ARGS for DecoderLSTM.__init__(hidden_size, num_layer=1)
INIT_ARGS = {
    'hidden_size': 8,
    'num_layer': 2
}

# ARGS for DecoderLSTM.forward(input_data, hidden)
# input_data shape: (seq_len, batch_size, hidden_size)
# hidden is a tuple of (h_0, c_0) each with shape (num_layer, batch_size, hidden_size)
batch_size = 3
seq_len = 5
hidden_size = INIT_ARGS['hidden_size']
num_layer = INIT_ARGS['num_layer']

ARGS = {
    'input_data': torch.randn(seq_len, batch_size, hidden_size),
    'hidden': (
        torch.randn(num_layer, batch_size, hidden_size),
        torch.randn(num_layer, batch_size, hidden_size)
    )
}
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 — a tuple output has no array shape of its own; none is 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
EncoderLSTM Ran
ran on a synthesized input
ApeironY/PCCSNet/PCCSNet/models/components.py
code served (permissive licence) · get_code("8cfa5370af11cd9b")
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. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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

class EncoderLSTM(nn.Module):

    def __init__(self, input_size, hidden_size, num_layer=3):
        super(EncoderLSTM, self).__init__()
        self.hidden_size = hidden_size
        self.num_layer = num_layer
        self.lstm = nn.LSTM(input_size, hidden_size, num_layer, bidirectional=True)

    def forward(self, input_data, hidden):
        output, hidden = self.lstm(input_data, hidden)
        return (output, hidden)

    def initHidden(self, batch_size):
        return (torch.zeros(self.num_layer * 2, batch_size, self.hidden_size).cuda(), torch.zeros(self.num_layer * 2, batch_size, self.hidden_size).cuda())


INIT_ARGS = {
    'input_size': 4,
    'hidden_size': 8,
    'num_layer': 2
}

ARGS = {
    'input_data': torch.randn(5, 2, 4),
    'hidden': (torch.zeros(4, 2, 8), torch.zeros(4, 2, 8))
}

if __name__ == '__main__':
    model = EncoderLSTM(**INIT_ARGS)
    output, hidden = model(**ARGS)
    print("Output shape:", output.shape)
    print("Hidden state shapes:", hidden[0].shape, hidden[1].shape)
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 — a tuple output has no array shape of its own; none is 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
PCCSNet Ran
ran on a synthesized input
ApeironY/PCCSNet/PCCSNet/models/components.py
code served (permissive licence) · get_code("4d53624d203a21f9")
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 Tensor, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
import torch.nn as nn

# Minimal stand-in for DecoderLSTM, EncoderLSTM, Synthesizer, Classifier, Decoder
# (these are defined in the class code above)

INIT_ARGS = {
    'obs_len': 8,
    'pre_len': 12,
    'n_cluster': 200,
    'obs_hidden_size': 48,
    'pred_hidden_size': 48,
    'num_layer': 3,
}

ARGS = {
    'obs_encoded': torch.randn(1, 48),
    'k': 5,
    'batch_labels': None,
    'all_modalities': torch.randn(200, 96),
    'obs_feature_len': None,
}
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 Tensor, ndim 3, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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)
Synthesizer Ran
ran on a synthesized input
ApeironY/PCCSNet/PCCSNet/models/components.py
code served (permissive licence) · get_code("500cb1c39abba542")
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 Tensor, ndim 2, 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
import torch.nn as nn

# INIT_ARGS: parameters for Synthesizer.__init__
INIT_ARGS = {
    'obs_hidden_size': 6,
    'pred_hidden_size': 8
}

# ARGS: parameters for Synthesizer.forward
batch_size = 2
obs_hidden_size = INIT_ARGS['obs_hidden_size']
pred_hidden_size = INIT_ARGS['pred_hidden_size']

# obs_encoded: shape (batch_size, obs_hidden_size)
obs_encoded = torch.randn(batch_size, obs_hidden_size)

# labels: shape (batch_size,) with integer indices
labels = torch.tensor([0, 1])

# all_modalities: shape (num_classes, obs_hidden_size + pred_hidden_size)
# This is indexed by labels, so we need at least max(labels)+1 rows
num_classes = 3
all_modalities = torch.randn(num_classes, obs_hidden_size + pred_hidden_size)

ARGS = {
    'obs_encoded': obs_encoded,
    'labels': labels,
    'all_modalities': all_modalities
}
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 Tensor, ndim 2, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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)
Decoder Not yet run
unverified (harvested, no recorded run)
ApeironY/PCCSNet/PCCSNet/models/components.py
code served (permissive licence) · get_code("4395402bac64bc7c")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 6 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.

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 prediction results are essential for trajectory prediction task as there is no single correct answer for the future. Previous frameworks can be divided into three categories: regression, generation and classification frameworks. However, these frameworks have weaknesses in different aspects so that they cannot model the multimodal prediction task comprehensively. In this paper, we present a novel insight along with a brand-new prediction framework by formulating multimodal prediction into three steps: modality clustering, classification and synthesis, and address the shortcomings of earlier frameworks. Exhaustive experiments on popular benchmarks have demonstrated that our proposed method surpasses state-of-the-art works even without introducing social and map information. Specifically, we achieve 19.2% and 20.8% improvement on ADE and FDE respectively on ETH/UCY dataset.

For agents

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

get_harvested_code_for_paper("2103.07854")
get_code_for_paper("2103.07854")
have("2103.07854")

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

curl https://syntology.ai/api/ran/2103.07854.json

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

[![Syntology run record](https://syntology.ai/api/ran/2103.07854.svg)](https://syntology.ai/paper/2103.07854)

Connect an agent — have() is free.