Papers › TrojanPuzzle: Covertly Poisoning Code-Suggestion Models

TrojanPuzzle: Covertly Poisoning Code-Suggestion Models

6 Jan 2023arXiv:2301.02344archive 2025-07-28

Hojjat Aghakhani, Wei Dai, Andre Manoel, Xavier Fernandes, Anant Kharkar, Christopher Kruegel, Giovanni Vigna, David Evans, Ben Zorn, Robert Sim

With tools like GitHub Copilot, automatic code suggestion is no longer a dream in software engineering. These tools, based on large language models, are typically trained on massive corpora of code mined from unvetted public sources. As a result, these models are susceptible to data poisoning attacks where an adversary manipulates the model's training by injecting malicious data. Poisoning attacks could be designed to influence the model's suggestions at run time for chosen contexts, such as inducing the model into suggesting insecure code payloads. To achieve this, prior attacks explicitly inject the insecure code payload into the training data, making the poison data detectable by static analysis tools that can remove such malicious data from the training set. In this work, we demonstrate two novel attacks, COVERT and TROJANPUZZLE, that can bypass static analysis by planting malicious poison data in out-of-context regions such as docstrings. Our most novel attack, TROJANPUZZLE, goes one step further in generating less suspicious poison data by never explicitly including certain (suspicious) parts of the payload in the poison data, while still inducing a model that suggests the entire payload when completing code (i.e., outside docstrings). This makes TROJANPUZZLE robust against signature-based dataset-cleansing methods that can filter out suspicious sequences from the training data. Our evaluation against models of two sizes demonstrates that both COVERT and TROJANPUZZLE have significant implications for practitioners when selecting code used to train or tune code-suggestion models.

PaperPDFCodeCode Syntology ran

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.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 13 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 13 samples from 1 repository, 10 ran (10 with no instrument failure; 0 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.

Levels written on 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

microsoft/codegenerationpoisoning officialmentioned in paperMITSyntology: 13 samples, 10 ran (0 instrument-failed) · 3 unverified report

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.

Code Syntology ran Syntology

13 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 3 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.

10ran
3unverified

Licence: 0 of the 13 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 microsoft/codegenerationpoisoning. “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.

microsoft/codegenerationpoisoning official repository 13 samples, 10 ran (0 instrument-failed) · 3 unverified
collator in 16 papers SalesforceCodeGen/training/fine_tune.py ran MIT (permissive) · 0bbeee52d85f7d22 · report fixture: signature data:list/1/tuple · returned dict · level recorded 2026-09-25
find in 13 papers analysis/collect_results.py ran fingerprinted MIT (permissive) · 2664dc3f210cc82b · report fixture: signature regex:str/0/str, s:str/0/str · returned str · level recorded 2026-09-25
find_ast_function baseline_attack.py ran MIT (permissive) · f07d13e8761a80a1 · report fixture: signature sample:str/0/str, payload:str/0/str, raise_error_if_no_func_found:bool/0/bool · returned FunctionDef · level recorded 2026-09-25
get_method_df analysis/barplot.py ran MIT (permissive) · a1505fe190179d58 · report fixture: signature df:DataFrame/0/?, method:str/0/str · returned DataFrame · level recorded 2026-09-25
get_orig_code_blocks context_agnostic_attack.py ran fingerprinted MIT (permissive) · fb9f7fa1d82df0ab · report fixture: signature code:str/0/str · returned list · level recorded 2026-09-25
if_compiles baseline_attack.py ran MIT (permissive) · 1f6d6c07dc4ebf58 · report fixture: signature f:str/0/str, do_print:bool/0/bool · returned bool · level recorded 2026-09-25
insert_trigger_randomly context_agnostic_attack.py ran MIT (permissive) · 4df03824e84c192d · report fixture: signature code:str/0/str, trigger:str/0/str · returned str · level recorded 2026-09-25
performance_from_completion_files analysis/collect_results.py ran MIT (permissive) · ba667370471940ec · report fixture: signature prompts_root:PosixPath/0/?, debugging:bool/0/bool · returned tuple · level recorded 2026-09-25
read_files in 18 papers baseline_attack.py ran MIT (permissive) · 2943ca9b8f035af3 · report fixture: signature files_dir:PosixPath/0/?, num:int/0/int · returned tuple · level recorded 2026-09-25
read_files in 18 papers context_agnostic_attack.py ran MIT (permissive) · 8fc4d487bec3de3d · report fixture: signature files_dir:PosixPath/0/?, num:int/0/int, max_len:int/0/int, tokenizer:callable/0/? · returned tuple · level recorded 2026-09-25
attack in 16 papers find_adversarial_docstring.py unverified MIT (permissive) · e3984b894794e44c · report
extract_config_from_path analysis/collect_results.py unverified MIT (permissive) · 201fc8201cfaf6b8 · report
gen in 18 papers SalesforceCodeGen/training/fine_tune.py unverified MIT (permissive) · ba8d70941078ed2f · report

Tasks

Data Poisoning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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