Papers › Jatmo: Prompt Injection Defense by Task-Specific Finetuning

Jatmo: Prompt Injection Defense by Task-Specific Finetuning

29 Dec 2023arXiv:2312.17673archive 2025-07-28

Julien Piet, Maha Alrashed, Chawin Sitawarin, Sizhe Chen, Zeming Wei, Elizabeth Sun, Basel Alomair, David Wagner

Large Language Models (LLMs) are attracting significant research attention due to their instruction-following abilities, allowing users and developers to leverage LLMs for a variety of tasks. However, LLMs are vulnerable to prompt-injection attacks: a class of attacks that hijack the model's instruction-following abilities, changing responses to prompts to undesired, possibly malicious ones. In this work, we introduce Jatmo, a method for generating task-specific models resilient to prompt-injection attacks. Jatmo leverages the fact that LLMs can only follow instructions once they have undergone instruction tuning. It harnesses a teacher instruction-tuned model to generate a task-specific dataset, which is then used to fine-tune a base model (i.e., a non-instruction-tuned model). Jatmo only needs a task prompt and a dataset of inputs for the task: it uses the teacher model to generate outputs. For situations with no pre-existing datasets, Jatmo can use a single example, or in some cases none at all, to produce a fully synthetic dataset. Our experiments on seven tasks show that Jatmo models provide similar quality of outputs on their specific task as standard LLMs, while being resilient to prompt injections. The best attacks succeeded in less than 0.5% of cases against our models, versus 87% success rate against GPT-3.5-Turbo. We release Jatmo at https://github.com/wagner-group/prompt-injection-defense.

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 18 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 8 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 18 samples from 1 repository, 10 ran (7 with no instrument failure; 3 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 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

wagner-group/prompt-injection-defense officialmentioned in paperSyntology: 18 samples, 10 ran (3 instrument-failed) · 8 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

18 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 8 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · our draft was wrong
7ran
8unverified

Licence: 18 of the 18 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 wagner-group/prompt-injection-defense. “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”. For 1 sample that ran, the graph records a signature this page does not show: an argument's recorded type or element kind is a printout of an object in Syntology's harness (such as a mock object, a class or a pandas printout), not the type of an input, so it does not describe the call's shape. “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.

wagner-group/prompt-injection-defense official repository 18 samples, 10 ran (3 instrument-failed) · 8 unverified
call_openai in 6 papers src/jatmo/server.py ran no licence file found · pointer only · 58f375594c3a65c3 · report fixture: returned Mock · level recorded 2026-09-24
format_finetune_data src/jatmo/automatic_pipeline/utils.py ran no licence file found · pointer only · afdfabcae731b1b3 · report fixture: signature inputs:list/1/str, outputs:list/1/str · returned list · level recorded 2026-09-24
gather_inputs src/jatmo/example_tasks/auto_tasks/code_summarization.py ran no licence file found · pointer only · 011dc1d185aadcee · report fixture: signature total_count:int/0/int, path:str/0/str · returned list · level recorded 2026-09-24
get_formatting_input src/jatmo/automatic_pipeline/utils.py ran no licence file found · pointer only · d8dd0fab84fbf910 · report fixture: signature task_description:str/0/str, inputs:str/0/str, example:tuple/1/str · returned tuple · level recorded 2026-09-24
get_generation_prompt src/jatmo/automatic_pipeline/utils.py ran no licence file found · pointer only · e0da5c3f437fc507 · report fixture: signature index:int/0/int, task:str/0/str, additional_rules:list/1/str, example:str/0/str, random_seed:int/0/int · returned tuple · level recorded 2026-09-24
init_servers src/jatmo/server.py ran · our draft was wrong no licence file found · pointer only · 57b9140b8e2e42bb · report fixture: signature number_of_processes:int/0/int · returned tuple · level recorded 2026-09-24
load_hackaprompt_injections src/jatmo/semi_automatic_pipeline/utils.py ran · our draft was wrong no licence file found · pointer only · af2d33fbd2f2574e · report fixture: signature path:str/0/str · returned list · level recorded 2026-09-24
normalize_book src/jatmo/example_tasks/auto_tasks/translation.py ran · our draft was wrong fingerprinted no licence file found · pointer only · de6f63f3c4a1f298 · report fixture: signature book:str/0/str · returned str · level recorded 2026-09-24
perturb_passage src/jatmo/semi_automatic_pipeline/utils.py ran fingerprinted no licence file found · pointer only · cf548b8291737d65 · report fixture: signature passage:str/0/str, position:int/0/int, injection:str/0/str · returned str · level recorded 2026-09-24
write_prompt src/jatmo/example_tasks/auto_tasks/review_summarization.py ran no licence file found · pointer only · 09ca1d2547f6362b · report fixture: signature reviews:list/1/str, position:NoneType/0/?, injection:NoneType/0/? · returned str · level recorded 2026-09-24
compare_to_ft_model src/jatmo/automatic_pipeline/eval_model.py unverified no licence file found · pointer only · f64affb6add891aa · report
format_inputs in 5 papers src/jatmo/automatic_pipeline/input_generation.py unverified no licence file found · pointer only · 4877041f759749d1 · report
gather_inputs src/jatmo/example_tasks/auto_tasks/news_summarization.py unverified no licence file found · pointer only · aa55ccc6d4abdf48 · report
gather_inputs src/jatmo/example_tasks/auto_tasks/review_summarization.py unverified no licence file found · pointer only · da6bcc46ae271b58 · report
gather_inputs src/jatmo/example_tasks/auto_tasks/translation.py unverified no licence file found · pointer only · 64875c7d8e21226c · report
get_input_list src/jatmo/automatic_pipeline/input_generation.py unverified no licence file found · pointer only · ecbd4b9dae46b569 · report
load_reviews src/jatmo/example_tasks/auto_tasks/review_summarization.py unverified no licence file found · pointer only · 45f097951892870c · report
rate_completions src/jatmo/server.py unverified no licence file found · pointer only · 3b5261f8ca59d546 · report

Tasks

Instruction Following

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AdamAttentionAttention DropoutBASEBPECosine AnnealingDense ConnectionsDropoutGPT-3Layer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionResidual ConnectionSoftmaxWeight Decay

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