Papers › A Watermark for Large Language Models
A Watermark for Large Language Models
John Kirchenbauer, Jonas Geiping, Yuxin Wen, Jonathan Katz, Ian Miers, Tom Goldstein
Potential harms of large language models can be mitigated by watermarking model output, i.e., embedding signals into generated text that are invisible to humans but algorithmically detectable from a short span of tokens. We propose a watermarking framework for proprietary language models. The watermark can be embedded with negligible impact on text quality, and can be detected using an efficient open-source algorithm without access to the language model API or parameters. The watermark works by selecting a randomized set of "green" tokens before a word is generated, and then softly promoting use of green tokens during sampling. We propose a statistical test for detecting the watermark with interpretable p-values, and derive an information-theoretic framework for analyzing the sensitivity of the watermark. We test the watermark using a multi-billion parameter model from the Open Pretrained Transformer (OPT) family, and discuss robustness and security.
In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.
Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Syntology holds a reference implementation of watermarking framework, generated from the paper text (not the authors' code), at level 3 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” (quoted from Syntology's server card). For this implementation, level 3 is recorded, but its report lists no property test that passed on both. No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='watermarking framework') (how to connect); the free check have('2301.10226') needs no token.
For agents, on Syntology's MCP service (how to connect):
get_harvested_code_for_paper(arxiv_id="2301.10226")get_code(code_sha256="…")with a sample'scode_sha256from the list belowget_citation_path(paper_1="2301.10226", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="2301.10226")
Code
Syntology 16 code samples harvested from 7 repositories linked to this paper: 11 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 1 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.
By repository: official repository: 3 samples from 1 repository, 2 ran (2 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 13 samples from 6 repositories, 9 ran (8 with no instrument failure; 1 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.
No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:
- huggingface/text-generation-inference, community (archive-listed): harvested for another paper, listed under the repository
Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 11 samples that ran has no recorded date.
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.
Code Syntology ran Syntology
16 samples harvested: 11 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 1 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.
Licence: 5 of the 16 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.
Harvested from 7 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.
Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.
Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.
A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.
63e6607939ea84a0 · report
fixture: signature input_ids:Tensor/2/int, scores:Tensor/2/float · returned Tensor · level recorded 2026-09-24 678874b625786116 · report
fixture: signature v:str/0/str · returned bool · level recorded, date not recorded bdea8c3520cb2de0 · report
fixture: signature tensor:Tensor/2/float · returned int · level recorded 2026-09-22 census: 5 implementations from 6 papers, 5 distinct outputs across 4 buckets, one shared input per bucket this sample: bucket 3 of 4, arg 1: rank 2, kind int, dtype int64 src/watermarking.py ran · our draft was wrong MIT (permissive) ·
e8a3bc02d0c28ef0 · report
fixture: signature model:callable/0/?, prior_tokens:Tensor/2/int, max_length:int/0/int, watermark:bool/0/bool, gamma:float/0/float, delta:int/0/int, hash_function:callable/0/? · returned Tensor · level recorded 2026-09-22 f26efcccf5a3772f · report
fixture: returned ContextMasking · level recorded 2026-09-22 42f6923bf0733dbe · report
fixture: returned CouplingGenerator · level recorded 2026-09-22 b773c77757980226 · report
fixture: returned OpenaiGenerator · level recorded 2026-09-22 1df0136501003b60 · report
fixture: returned WmGenerator · level recorded 2026-09-22 19f3e32f390a20ae · report
51fae2306a5ebeda · report
13b864aa60fdbcdd · report
fixture: returned WmGenerator · level recorded 2026-09-22 e4a6e1de65880ce7 · report
fixture: returned MarylandGenerator · level recorded 2026-09-22 84002164ce29edb7 · report
fixture: returned WmGenerator · level recorded 2026-09-22 dd584304f6d90615 · report
a74f13e55a5c37a6 · report
Method as code Syntology
What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another. Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.
8 assemblies on this paper: 7 harvested, 1 generated, listed in 2 groups (siblings lifted in one batch form one group).
generated Assembly for watermarking framework (graph method: watermarking framework), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer; level 3 is recorded, but its report lists no property test that passed on both.
label matches by construction: the code was generated for this method (not judged by a model)
- compute_greenlist_decision entry · level 3 · ran · our draft was wrong ·
5a8a21456e78908a0e0bbb76f30d9c4c866e07e0fb3c13d91b81d1a928d85639
7 sibling assemblies for watermarking framework (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.
harvested Assembly for watermarking framework (graph method: watermarking framework). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.
one of 7 siblings
- WmGenerator entry · level 1 · ran ·
13b864aa60fdbcdd
harvested Assembly for watermarking framework (graph method: watermarking framework). 2 bricks; individually on a fixture: 2 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; every one of the 2 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.
one of 7 siblings
- MarylandGenerator entry · level 1 · ran ·
e4a6e1de65880ce7 - WmGenerator helper · level 1 · ran ·
84002164ce29edb7
harvested Assembly for watermarking framework (graph method: watermarking framework). 4 bricks; individually on a fixture: 4 ran (of which 4 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; every one of the 4 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.
one of 7 siblings
- CouplingGenerator entry · level 1 · ran ·
42f6923bf0733dbe - OpenaiGenerator helper · level 1 · ran ·
b773c77757980226 - WmGenerator helper · level 1 · ran ·
1df0136501003b60 - ContextMasking helper · level 1 · ran ·
f26efcccf5a3772f
harvested Assembly for watermarking framework (graph method: watermarking framework). 2 bricks; individually on a fixture: 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.
one of 7 siblings
- generate entry · level 1 · ran · our draft was wrong ·
e8a3bc02d0c28ef0 - default_hash_fn helper · level 1 · ran · honoured contract ·
bdea8c3520cb2de0
harvested Assembly for watermarking framework (graph method: watermarking framework). 2 bricks; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.
one of 7 siblings
- WatermarkLogitsProcessor entry · level 1 · ran ·
63e6607939ea84a0 - WatermarkBase helper · level 0 · unverified ·
ba3557aac9e7bf4c
harvested Assembly for watermarking framework (graph method: watermarking framework). 2 bricks; individually on a fixture: 0 ran · 2 unverified. The assembly has not been run end-to-end.
one of 7 siblings
- WatermarkLogitsProcessor entry · level 0 · unverified ·
a74f13e55a5c37a6 - WatermarkBase helper · level 0 · unverified ·
dd584304f6d90615
harvested Assembly for watermarking framework (graph method: watermarking framework). 2 bricks; individually on a fixture: 0 ran · 2 unverified. The assembly has not been run end-to-end.
one of 7 siblings
- WatermarkLogitsProcessor entry · level 0 · unverified ·
51fae2306a5ebeda - WatermarkBase helper · level 0 · unverified ·
19f3e32f390a20ae
Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2301.10226") returns these assemblies with each brick's role and wiring (how to connect).
Tasks
Results from the paper archive 2025-07-28
No leaderboard rows for this paper in the archive.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections