SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.04411 · 2024

Waterfall: Framework for Robust and Scalable Text Watermarking and Provenance for LLMs

arXiv · PDF · Open in the Atlas

Code that ran

1 of 2 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 0 have no contract verdict; 1 failure 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/2407.04411.json · badge: /api/ran/2407.04411.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
aoi3142/waterfall canonical (classifier label)
classifier basis: census
1 ran · 1 no recorded run
FunctionStatusWhere it lives
indices_to_counts Ran
ran; our contract draft was wrong, not the code
aoi3142/waterfall/waterfall/WatermarkerBase.py
code served (permissive licence) · get_code("41c26cf8365ff2a3")
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 csr_matrix. 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 numpy as np
from scipy.sparse import csr_matrix

# Construct arguments for indices_to_counts function
# N: int - the number of columns in the output sparse matrix
# dtype: np.dtype - the data type for the bincount output
# indices: np.ndarray - a 2D array where each row contains indices to count

N = 5
dtype = np.int32

# Create a 2D array of indices
# Each row will be processed independently by bincount
# Using values in range [0, N)
indices = np.array([
    [0, 1, 2, 1, 0],
    [3, 4, 2, 0, 1]
], dtype=np.int64)

ARGS = {
    "N": N,
    "dtype": dtype,
    "indices": indices
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type csr_matrix — a csr_matrix output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a csr_matrix 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
generate Not yet run
unverified (harvested, no recorded run)
aoi3142/waterfall/waterfall/group-beam-search/custom_generate/generate.py
code served (permissive licence) · get_code("e82bfea8b311483b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 2 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 0 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
generate 1 other paper
1 same year, order not determinable
2024.emnlp-main.1138 (2024 (year only))
Waterfall: Scalable Framework for Robust Text Watermarking and Provenance for LLMs
indices_to_counts 1 other paper
1 same year, order not determinable
2024.emnlp-main.1138 (2024 (year only))
Waterfall: Scalable Framework for Robust Text Watermarking and Provenance for LLMs

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

Protecting intellectual property (IP) of text such as articles and code is increasingly important, especially as sophisticated attacks become possible, such as paraphrasing by large language models (LLMs) or even unauthorized training of LLMs on copyrighted text to infringe such IP. However, existing text watermarking methods are not robust enough against such attacks nor scalable to millions of users for practical implementation. In this paper, we propose Waterfall, the first training-free framework for robust and scalable text watermarking applicable across multiple text types (e.g., articles, code) and languages supportable by LLMs, for general text and LLM data provenance. Waterfall comprises several key innovations, such as being the first to use LLM as paraphrasers for watermarking along with a novel combination of techniques that are surprisingly effective in achieving robust verifiability and scalability. We empirically demonstrate that Waterfall achieves significantly better scalability, robust verifiability, and computational efficiency compared to SOTA article-text watermarking methods, and showed how it could be directly applied to the watermarking of code. We also demonstrated that Waterfall can be used for LLM data provenance, where the watermarks of LLM training data can be detected in LLM output, allowing for detection of unauthorized use of data for LLM training and potentially enabling model-centric watermarking of open-sourced LLMs which has been a limitation of existing LLM watermarking works. Our code is available at https://github.com/aoi3142/Waterfall.

For agents

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

get_harvested_code_for_paper("2407.04411")
get_code_for_paper("2407.04411")
have("2407.04411")

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

curl https://syntology.ai/api/ran/2407.04411.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.04411.svg)](https://syntology.ai/paper/2407.04411)

Connect an agent — have() is free.