SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.23402 · ICML · 2026

Parametric Prior Mapping Framework for Non-stationary Probabilistic Time Series Forecasting

Ning Gui, Jinglin Li, Jun Tan, Q Fang

arXiv · PDF · Open in the Atlas

Code that ran

4 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 3 failures of our instrument; 4 no recorded run), measured 2026-11-21; 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/2605.23402.json · badge: /api/ran/2605.23402.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
ljl8336/PPM community
Papers with Code archive
3 ran · 4 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
FunctionStatusWhere it lives
GrowConnections Ran
ran on a synthesized input · constructed an object; no result was computed
ljl8336/PPM/layers/SVQ/SVQ_block.py
code served (permissive licence) · get_code("1687769c46951959")
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 an instance of its own class, GrowConnections: the run constructed the object and did not call it, so no result was computed. 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 math

torch.manual_seed(0)

# Set up tensor dimensions
# in_features: input feature dimension
# out_features: output feature dimension
# batch_size: number of samples
in_features = 3
out_features = 4
batch_size = 2

# Create inputs tensor: shape (batch_size, in_features)
inputs = torch.randn(batch_size, in_features)

# Create a sparse weight matrix represented by indices and weights
# indices: shape (2, num_connections) where first row is output indices, second row is input indices
# k: number of new connections to grow
k = 2

# Create valid sparse indices (out_idx, in_idx pairs)
num_connections = 5
indices = torch.randint(0, min(out_features, in_features), (2, num_connections))
# Ensure unique indices for sparse tensor validity
indices = torch.unique(indices, dim=1)
if indices.shape[1] < num_connections:
    # Pad with valid indices
    remaining = num_connections - indices.shape[1]
    extra_indices = torch.randint(0, min(out_features, in_features), (2, remaining))
    indices = torch.cat([indices, extra_indices], dim=1)

# Create weights for the sparse connections
weights = torch.randn(indices.shape[1])

# features tuple: (out_features, in_features)
features = (out_features, in_features)

# max_size: maximum size of the weight matrix (used for memory-efficient processing)
max_size = 100

ARGS = {
    'inputs': inputs,
    'weights': weights,
    'k': k,
    'indices': indices,
    'features': features,
    'max_size': max_size
}
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 GrowConnections — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GrowConnections 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
gumbel_sample Ran
ran; our contract draft was wrong, not the code
ljl8336/PPM/layers/SVQ/SVQ_block.py
code served (permissive licence) · get_code("920615834753c0f1")
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 tuple. 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.functional as F

torch.manual_seed(0)

# Define the helper functions that gumbel_sample depends on
def log(t, eps=1e-20):
    return torch.log(t.clamp(min=eps))

def gumbel_noise(t):
    noise = torch.zeros_like(t).uniform_(0, 1)
    return -log(-log(noise))

# Create a small logits tensor: batch_size=2, num_classes=4
logits = torch.randn(2, 4)

# Define ARGS dict with all parameters
ARGS = {
    'logits': logits,
    'temperature': 1.0,
    'stochastic': True,
    'straight_through': False,
    'reinmax': False,
    'dim': -1,
    'training': True
}
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 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: none by the harness; any seed is in the setup block
kmeans Ran
ran; our contract draft was wrong, not the code
ljl8336/PPM/layers/SVQ/SVQ_block.py
code served (permissive licence) · get_code("e139cdc24aa46838")
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 tuple. 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.functional as F
from einops import rearrange, repeat, reduce, pack, unpack

torch.manual_seed(0)

# Create small sample tensors
# Shape: (num_codebooks, num_samples, dim)
num_codebooks = 2
num_samples = 5
dim = 4
samples = torch.randn(num_codebooks, num_samples, dim)

# Number of clusters
num_clusters = 3

# Number of iterations
num_iters = 2

# Use cosine similarity flag
use_cosine_sim = False

# Sample function - use the provided batched_sample_vectors
def batched_sample_vectors(samples, num):
    def sample_vectors(samples, num):
        num_samples, device = (samples.shape[0], samples.device)
        if num_samples >= num:
            indices = torch.randperm(num_samples, device=device)[:num]
        else:
            indices = torch.randint(0, num_samples, (num,), device=device)
        return samples[indices]
    
    return torch.stack([sample_vectors(sample, num) for sample in samples.unbind(dim=0)], dim=0)

sample_fn = batched_sample_vectors

# All reduce function - use noop
def noop(*args, **kwargs):
    pass

all_reduce_fn = noop

ARGS = {
    'samples': samples,
    'num_clusters': num_clusters,
    'num_iters': num_iters,
    'use_cosine_sim': use_cosine_sim,
    'sample_fn': sample_fn,
    'all_reduce_fn': all_reduce_fn,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, 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: none by the harness; any seed is in the setup block
small_world_chunker Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
this paper's copy was not recorded; identical code first harvested from hyeon95y/sparselinear
pointer only · get_code("3a0749baa452d466")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

CosineSimCodebook Not yet run
unverified (harvested, no recorded run)
ljl8336/PPM/layers/SVQ/SVQ_block.py
code served (permissive licence) · get_code("4e1520083d05dd61")
How it ran

no invocation recorded.

EuclideanCodebook Not yet run
unverified (harvested, no recorded run)
ljl8336/PPM/layers/SVQ/SVQ_block.py
code served (permissive licence) · get_code("d57ad3d13bb25601")
How it ran

no invocation recorded.

SparseLinear Not yet run
unverified (harvested, no recorded run)
ljl8336/PPM/layers/SVQ/SVQ_block.py
code served (permissive licence) · get_code("8329d0d60dc6cf18")
How it ran

no invocation recorded.

VectorQuantize Not yet run
unverified (harvested, no recorded run)
ljl8336/PPM/layers/SVQ/SVQ_block.py
code served (permissive licence) · get_code("72502b0faf46a86f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 0 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
kmeans
brick of this paper's assembly: helper
3 other papers
3 dated earlier · 1 only as an assembly brick
2308.02117 (2023-08)
VQGraph: Rethinking Graph Representation Space for Bridging GNNs and MLPs
2311.17245 (2023-11)
LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPS
2402.16321 (2024-02)
Self-Supervised Speech Quality Estimation and Enhancement Using Only Clean Speech
small_world_chunker
brick of this paper's assembly: helper
2 other papers
2 dated earlier
1903.11257 (2019-03)
How Can We Be So Dense? The Benefits of Using Highly Sparse Representations
1911.11134 (2019-11)
Rigging the Lottery: Making All Tickets Winners
gumbel_sample
brick of this paper's assembly: sampler
1 other paper
1 dated earlier
2411.06070 (2024-11)
GFT: Graph Foundation Model with Transferable Tree Vocabulary

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Effectively modeling non-stationary dynamics in probabilistic multivariate time series(MTS) forecasting requires balancing expressiveness with robustness. Existing parametric approaches benefit from strong inductive biases but lack flexibility, whereas deep generative models struggle to capture complex temporal dependencies without extensive data and computation. We introduce Parametric Prior Mapping (PPM), a framework that injects parametric structural priors into a generative modeling process. Specifically, PPM utilizes a parametric estimator to derive a dynamic, adaptive prior that guides the learning of a complex predictive distribution via a learnable mapping. This design allows the model to retain the efficiency of parametric methods while exploiting the expressive power of generative models. Trained with a hybrid objective, PPM yields precise forecasts with well-calibrated uncertainty estimates. Empirical results show that PPM outperforms existing baselines in handling non-stationary data, offering a superior trade-off between accuracy and computational efficiency. The code is available at https://github.com/ljl8336/PPM.

For agents

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

get_harvested_code_for_paper("2605.23402")
get_code_for_paper("2605.23402")
have("2605.23402")

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

curl https://syntology.ai/api/ran/2605.23402.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.23402.svg)](https://syntology.ai/paper/2605.23402)

Connect an agent — have() is free.