SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.16187 · 2024

No Free Lunch in LLM Watermarking: Trade-offs in Watermarking Design Choices

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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, 1 violated it and 0 have no contract verdict; 2 failures of our instrument; 0 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/2402.16187.json · badge: /api/ran/2402.16187.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
qi-pang/llm-watermark-attacks canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
FunctionStatusWhere it lives
read_file Ran
ran; our contract draft was wrong, not the code
qi-pang/llm-watermark-attacks/Exp-Watermark/experiments/attack_multiple_keys.py
code served (permissive licence) · get_code("fc9d1d8c92c8dd16")
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, 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
import json
import tempfile
import os

# Create a temporary file with JSON lines
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.jsonl")

# Write some test JSON lines to the file
test_data = [
    {"key1": "value1", "key2": 42},
    {"key1": "value2", "key2": 43},
    {"key1": "value3", "key2": 44}
]

with open(temp_file, 'w') as f:
    for item in test_data:
        f.write(json.dumps(item) + '\n')

ARGS = {
    "filename": temp_file
}
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, ndim 1, dtype family float — a list 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: none by the harness; any seed is in the setup block
read_file Ran
ran; our contract draft was wrong, not the code
qi-pang/llm-watermark-attacks/KGW-Watermark/attack_multiple_keys.py
code served (permissive licence) · get_code("48e3a03b6b0feb73")
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, 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 os
import json
import tempfile

# Create a temporary file with JSON lines
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.jsonl")

# Write some test data
test_data = [
    {"key1": "value1", "key2": 42},
    {"key1": "value2", "key2": 43},
    {"key1": "value3", "key2": 44}
]

with open(temp_file, 'w') as f:
    for item in test_data:
        f.write(json.dumps(item) + '\n')

ARGS = {
    "filename": temp_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list 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: none by the harness; any seed is in the setup block
str2bool Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from jwkirchenbauer/lm-watermarking
pointer only · get_code("678874b625786116")
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.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 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 (2 with at least one paper dated earlier, 1 the same month, 1 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
str2bool 175 other papers
104 dated earlier, 8 the same month, 62 later, 1 same year, order not determinable · 2 only as an assembly brick
1406.6247 (2014-06)
Recurrent Models of Visual Attention
1603.09382 (2016-03)
Deep Networks with Stochastic Depth
1710.10196 (2017-10)
Progressive Growing of GANs for Improved Quality, Stability, and Variation
and 172 more not named here
read_file 1 other paper
1 dated earlier
2306.17439 (2023-06)
Provable Robust Watermarking for AI-Generated Text

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

Advances in generative models have made it possible for AI-generated text, code, and images to mirror human-generated content in many applications. Watermarking, a technique that aims to embed information in the output of a model to verify its source, is useful for mitigating the misuse of such AI-generated content. However, we show that common design choices in LLM watermarking schemes make the resulting systems surprisingly susceptible to attack -- leading to fundamental trade-offs in robustness, utility, and usability. To navigate these trade-offs, we rigorously study a set of simple yet effective attacks on common watermarking systems, and propose guidelines and defenses for LLM watermarking in practice.

For agents

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

get_harvested_code_for_paper("2402.16187")
get_code_for_paper("2402.16187")
have("2402.16187")

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

curl https://syntology.ai/api/ran/2402.16187.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.16187.svg)](https://syntology.ai/paper/2402.16187)

Connect an agent — have() is free.