SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2307.09306 · ICCV · 2023

EigenTrajectory: Low-Rank Descriptors for Multi-Modal Trajectory Forecasting

Jean Oh, Hae-Gon Jeon, Inhwan Bae

arXiv · PDF · Open in the Atlas

Code that ran

7 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; 2 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 4 failures of our instrument; 2 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/2307.09306.json · badge: /api/ran/2307.09306.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
inhwanbae/EigenTrajectory canonical (classifier label)
classifier basis: census
7 ran · 2 no recorded run
FunctionStatusWhere it lives
curve_fitting Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
inhwanbae/EigenTrajectory/CurveModel/curve_fitting.py
code served (permissive licence) · get_code("ffc38fcf2686c808")
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 3, dtype family float, shape (2, 5, 3) as a later run recorded it. 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, 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 sys

torch.manual_seed(0)

# Create trajectory tensor: (n_ped, t_traj, dim)
# n_ped=2 pedestrians, t_traj=5 time steps, dim=3 spatial dimensions
traj = torch.randn(2, 5, 3, dtype=torch.float32)

# Create basis tensor: (t_traj, n_cp)
# t_traj=5 time steps, n_cp=4 control points
basis = torch.randn(5, 4, dtype=torch.float32)

ARGS = {
    'traj': traj,
    'basis': basis,
}
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 3, dtype family float, shape (2, 5, 3) — shape (2, 5, 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: none by the harness; any seed is in the setup block
generate_ar_mask Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
inhwanbae/EigenTrajectory/baseline/agentformer/model.py
code served (permissive licence) · get_code("1b373d997a81df48")
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 (6, 6) 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 numpy as np

torch.manual_seed(0)

# Analyze the function:
# - sz: int, must be divisible by agent_num
# - agent_num: int, divides sz
# - agent_mask: a tensor that supports .repeat(T, T) method
#   The function treats it as a 2D mask that gets repeated
#   Looking at usage: mask[i1:i2, i2:] = float('-inf')
#   So agent_mask should be a 2D tensor of shape (agent_num, agent_num)

# Create a small agent_mask tensor
# agent_num = 2, so agent_mask is (2, 2)
agent_num = 2
agent_mask = torch.ones(agent_num, agent_num)

# sz must be divisible by agent_num
# Let's use sz = 6, so T = 3
sz = 6

ARGS = {
    "sz": sz,
    "agent_num": agent_num,
    "agent_mask": agent_mask,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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 (6, 6) — shape (6, 6) 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
generate_identity_matrix Ran
ran; our contract draft was wrong, not the code
inhwanbae/EigenTrajectory/baseline/gpgraphsgcn/model_groupwrapper.py
code served (permissive licence) · get_code("828ed392be109d01")
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 list. 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

torch.manual_seed(0)

# The function expects a tensor v with at least 4 dimensions
# It accesses v.size(2), v.size(3), v.device
# It calls torch.eye() and .repeat() on the result
# Create a 4D tensor with distinct sizes for each dimension
v = torch.randn(2, 3, 4, 5)

ARGS = {
    "v": v
}
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 list — a list 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: none by the harness; any seed is in the setup block
generate_mask Ran
ran, violated the contract we drafted
inhwanbae/EigenTrajectory/baseline/agentformer/model.py
code served (permissive licence) · get_code("66d9c97d20cb3bfd")
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 (4, 9) 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 numpy as np

torch.manual_seed(0)

# Analyze the function:
# - tgt_sz: integer (target size)
# - src_sz: integer (source size)
# - agent_num: integer (number of agents)
# - agent_mask: a tensor that supports .repeat() method
#
# The function:
# 1. Asserts tgt_sz % agent_num == 0 and src_sz % agent_num == 0
# 2. Calls agent_mask.repeat(tgt_sz // agent_num, src_sz // agent_num)
# 3. Returns the result
#
# agent_mask needs to be a 2D tensor since repeat is called with 2 arguments

# Choose values that satisfy the divisibility constraints
# Use distinct sizes for clarity
tgt_sz = 8
src_sz = 6
agent_num = 2

# agent_mask should be a 2D boolean or numeric tensor
# Shape can be small, e.g., (1, 1) or (2, 3)
agent_mask = torch.tensor([[True, False, True]], dtype=torch.bool)

ARGS = {
    "tgt_sz": tgt_sz,
    "src_sz": src_sz,
    "agent_num": agent_num,
    "agent_mask": agent_mask
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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 (4, 9) — shape (4, 9) 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
irwin_hall_pdf Ran
ran, honoured the contract we drafted
inhwanbae/EigenTrajectory/CurveModel/curve_basis.py
code served (permissive licence) · get_code("e710d9e59161e4ba")
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 1, dtype family float, shape (1,) 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)

# Analyze the function signature and body:
# irwin_hall_pdf(n, x)
# - n: used in torch.arange(0, n + 1, 1, dtype=torch.float)
#      and in torch.ones(n + 1) * n
#      and in (x - k) ** (n - 1)
#      and in torch_pactorial(torch.FloatTensor([n]) - 1)
#   So n should be a scalar or tensor that can be used in arithmetic/indexing
# - x: used in (x - k) for broadcasting with k (which has shape n+1)
#      x - k must produce a tensor
#   x should be a scalar or tensor compatible with broadcasting

# Based on the function body, n appears to be used as a scalar value
# (converted to float for arange, used in arithmetic)
# x appears to be a scalar or small tensor

# Looking at how sgn = x - k is computed and then used:
# k has shape (n+1,) as a 1D tensor
# x - k should broadcast properly, so x should be a scalar

n = 3  # Small integer for Irwin-Hall distribution parameter
x = torch.tensor(1.5)  # Scalar tensor for the evaluation point

ARGS = {
    "n": n,
    "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 1, dtype family float, shape (1,) — shape (1,) 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
torch_binom Ran
ran, honoured the contract we drafted
inhwanbae/EigenTrajectory/CurveModel/curve_basis.py
code served (permissive licence) · get_code("fba974066d320431")
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)

# torch_binom expects two tensor arguments: n and k
# The function uses:
# - .detach() on both n and k
# - Comparison operations (>=)
# - Multiplication with mask
# - torch.lgamma() on n, n-k, and k
# - torch.exp()
# 
# Both n and k should be tensors with numeric values

n = torch.tensor([[5.0, 6.0, 7.0, 8.0],
                   [9.0, 10.0, 11.0, 12.0]], dtype=torch.float32)

k = torch.tensor([[2.0, 3.0, 1.0, 4.0],
                   [2.0, 5.0, 3.0, 2.0]], dtype=torch.float32)

ARGS = {
    'n': n,
    'k': k
}
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
torch_pactorial Ran
ran; our contract draft was wrong, not the code
inhwanbae/EigenTrajectory/CurveModel/curve_basis.py
code served (permissive licence) · get_code("298994d0e5c87a7f")
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 1, dtype family float, shape (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)

# torch_pactorial computes factorial using lgamma and exp
# It takes a single parameter n which should be a tensor or number
# The function calls torch.lgamma(n + 1).exp()

n = torch.tensor([1.0, 2.0, 3.0, 4.0])

ARGS = {
    "n": n
}
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 1, dtype family float, shape (4,) — shape (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
generate_adjacency_matrix Not yet run
unverified (harvested, no recorded run)
inhwanbae/EigenTrajectory/baseline/gpgraphsgcn/model_groupwrapper.py
code served (permissive licence) · get_code("18eeb7c1c4d68a72")
How it ran

no invocation recorded.

get_GPGraph_SGCN_model Not yet run
unverified (harvested, no recorded run)
inhwanbae/EigenTrajectory/baseline/gpgraphsgcn/model.py
code served (permissive licence) · get_code("3b78c4b8056c3cf9")
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, 4 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 1 later) and 5 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
torch_binom 2 other papers
2 later
2309.13167 (2023-09)
Flow Factorized Representation Learning
2410.05564 (2024-10)
Unsupervised Representation Learning from Sparse Transformation Analysis
generate_ar_mask 1 other paper
1 dated earlier
2103.14023 (2021-03)
AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting
generate_identity_matrix 1 other paper
1 dated earlier
2207.09953 (2022-07)
Learning Pedestrian Group Representations for Multi-modal Trajectory Prediction
generate_mask 1 other paper
1 dated earlier
2103.14023 (2021-03)
AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting

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

Capturing high-dimensional social interactions and feasible futures is essential for predicting trajectories. To address this complex nature, several attempts have been devoted to reducing the dimensionality of the output variables via parametric curve fitting such as the Bézier curve and Bspline function. However, these functions, which originate in computer graphics fields, are not suitable to account for socially acceptable human dynamics. In this paper, we present EigenTrajectory (ET), a trajectory prediction approach that uses a novel trajectory descriptor to form a compact space, known here as ET space, in place of Euclidean space, for representing pedestrian movements. We first reduce the complexity of the trajectory descriptor via a low-rank approximation. We transform the pedestrians' history paths into our ET space represented by spatio-temporal principle components, and feed them into off-the-shelf trajectory forecasting models. The inputs and outputs of the models as well as social interactions are all gathered and aggregated in the corresponding ET space. Lastly, we propose a trajectory anchor-based refinement method to cover all possible futures in the proposed ET space. Extensive experiments demonstrate that our EigenTrajectory predictor can significantly improve both the prediction accuracy and reliability of existing trajectory forecasting models on public benchmarks, indicating that the proposed descriptor is suited to represent pedestrian behaviors. Code is publicly available at https: //github.com/inhwanbae/EigenTrajectory.

For agents

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

get_harvested_code_for_paper("2307.09306")
get_code_for_paper("2307.09306")
have("2307.09306")

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

curl https://syntology.ai/api/ran/2307.09306.json

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

[![Syntology run record](https://syntology.ai/api/ran/2307.09306.svg)](https://syntology.ai/paper/2307.09306)

Connect an agent — have() is free.