Papers › Self-Instruct: Aligning Language Models with Self-Generated Instructions

Self-Instruct: Aligning Language Models with Self-Generated Instructions

20 Dec 2022arXiv:2212.10560archive 2025-07-28

Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A. Smith, Daniel Khashabi, Hannaneh Hajishirzi

Large "instruction-tuned" language models (i.e., finetuned to respond to instructions) have demonstrated a remarkable ability to generalize zero-shot to new tasks. Nevertheless, they depend heavily on human-written instruction data that is often limited in quantity, diversity, and creativity, therefore hindering the generality of the tuned model. We introduce Self-Instruct, a framework for improving the instruction-following capabilities of pretrained language models by bootstrapping off their own generations. Our pipeline generates instructions, input, and output samples from a language model, then filters invalid or similar ones before using them to finetune the original model. Applying our method to the vanilla GPT3, we demonstrate a 33% absolute improvement over the original model on Super-NaturalInstructions, on par with the performance of InstructGPT-001, which was trained with private user data and human annotations. For further evaluation, we curate a set of expert-written instructions for novel tasks, and show through human evaluation that tuning GPT3 with Self-Instruct outperforms using existing public instruction datasets by a large margin, leaving only a 5% absolute gap behind InstructGPT-001. Self-Instruct provides an almost annotation-free method for aligning pre-trained language models with instructions, and we release our large synthetic dataset to facilitate future studies on instruction tuning. Our code and data are available at https://github.com/yizhongw/self-instruct.

PaperPDFCodeCode Syntology ranMethod as code

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 17 code samples harvested from 6 repositories linked to this paper: 10 ran (of which 3 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 4 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 2 samples from 2 repositories, 2 ran (0 with no instrument failure; 2 where Syntology's instrument failed); community (archive-listed): 15 samples from 4 repositories, 8 ran (6 with no instrument failure; 2 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 10 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 10 samples that ran has no recorded date.

19 repositories listed; official and paper-mentioned ones first.

tatsu-lab/stanford_alpaca officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
yizhongw/self-instruct officialmentioned in papermentioned on GitHubSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
baai-zlab/coig mentioned on GitHubSyntology: not harvested report
beomi/koalpaca mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to LLaMA: Open and Efficient Foundation Language Models (arXiv:2302.13971; that paper's own run record: 37 ran (of which 9 constructed an object rather than computing a result; 25 with no instrument failure: 3 honoured, 0 violated, 22 with no contract checked; 12 where Syntology's instrument failed) · 21 unverified); nothing checks that they implement this paper's method. report
camel-ai/camel mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society (arXiv:2303.17760; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
daniel-furman/sft-demos mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to QLoRA: Efficient Finetuning of Quantized LLMs (arXiv:2305.14314; that paper's own run record: 18 ran (of which 1 constructed an object rather than computing a result; 6 with no instrument failure: 2 honoured, 2 violated, 2 with no contract checked; 12 where Syntology's instrument failed) · 8 unverified); nothing checks that they implement this paper's method. report
databrickslabs/dolly mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to XPERT: Expert Knowledge Transfer for Effective Training of Language Models (arXiv:2605.08842; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
facico/chinese-vicuna mentioned on GitHubpytorchSyntology: 3 samples, 2 ran (0 instrument-failed) · 1 unverified report
flagopen/flaginstruct mentioned on GitHubApache-2.0Syntology: not harvested report
fsoft-ai4code/codecapybara mentioned on GitHubpytorchSyntology: 5 samples, 2 ran (2 instrument-failed) · 3 unverified report
imoneoi/openchat mentioned on GitHubpytorchSyntology: 3 samples, 1 ran (0 instrument-failed) · 2 unverified report
jondurbin/airoboros mentioned on GitHubpytorchApache-2.0Syntology: 4 samples, 3 ran (0 instrument-failed) · 1 unverified report
lightaime/camel mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Scaling Synthetic Data Creation with 1,000,000,000 Personas (arXiv:2406.20094; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
ofa-sys/expertllama mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to LLaMA: Open and Efficient Foundation Language Models (arXiv:2302.13971; that paper's own run record: 37 ran (of which 9 constructed an object rather than computing a result; 25 with no instrument failure: 3 honoured, 0 violated, 22 with no contract checked; 12 where Syntology's instrument failed) · 21 unverified); nothing checks that they implement this paper's method. report
phoebussi/alpaca-cot mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to LoRA: Low-Rank Adaptation of Large Language Models (arXiv:2106.09685; that paper's own run record: 51 ran (of which 19 constructed an object rather than computing a result; 43 with no instrument failure: 1 honoured, 0 violated, 42 with no contract checked; 8 where Syntology's instrument failed) · 33 unverified) and ACL Anthology id 2023.findings-emnlp.269 (that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
replicate/cog_stanford_alpaca mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
sahil280114/codealpaca mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to SelfCodeAlign: Self-Alignment for Code Generation (arXiv:2410.24198; that paper's own run record: 30 ran (of which 3 constructed an object rather than computing a result; 21 with no instrument failure: 1 honoured, 0 violated, 20 with no contract checked; 9 where Syntology's instrument failed) · 7 unverified); nothing checks that they implement this paper's method. report
teelinsan/camoscio mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to LLaMA: Open and Efficient Foundation Language Models (arXiv:2302.13971; that paper's own run record: 37 ran (of which 9 constructed an object rather than computing a result; 25 with no instrument failure: 3 honoured, 0 violated, 22 with no contract checked; 12 where Syntology's instrument failed) · 21 unverified); nothing checks that they implement this paper's method. report
xzhang97666/alpacare mentioned on GitHubApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to AlpaCare:Instruction-tuned Large Language Models for Medical Application (arXiv:2310.14558; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. 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.

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

17 samples harvested: 10 ran (of which 3 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 4 where Syntology's instrument failed) · 7 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.

4ran · our draft was wrong
6ran
7unverified

Licence: 1 of the 17 samples is 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 6 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. 1 sample whose link to this paper records no repository is listed under one of this paper's own repositories, because identical code was harvested from it for another 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.

tatsu-lab/stanford_alpaca official repository 1 sample, 1 ran (1 instrument-failed) · 0 unverified
find_word_in_string in 12 papers generate_instruction.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · our draft was wrong fingerprinted licence of this copy not recorded · 8ca6a5f9f77053c2 · report fixture: signature w:str/0/str, s:str/0/str · returned Match · level recorded, date not recorded
yizhongw/self-instruct official repository 1 sample, 1 ran (1 instrument-failed) · 0 unverified
encode_prompt in 54 papers self_instruct/bootstrap_instructions.py ran · our draft was wrong Apache-2.0 (permissive) · e0456866b386ce58 · report fixture: signature prompt_instructions:list/1/str, classification:bool/0/bool · returned str · level recorded 2026-09-22
facico/chinese-vicuna community (archive-listed) 3 samples, 2 ran (0 instrument-failed) · 1 unverified
instruct_prompt prompt.py ran Apache-2.0 (permissive) · f56f83bd10a33749 · report fixture: returned instruct_prompt · level recorded 2026-09-22
prompt in 7 papers prompt.py ran Apache-2.0 (permissive) · 1171a8f3cf445159 · report fixture: returned prompt · level recorded 2026-09-22
printf prompt.py unverified Apache-2.0 (permissive) · f53b42abf295354b · report
fsoft-ai4code/codecapybara community (archive-listed) 5 samples, 2 ran (2 instrument-failed) · 3 unverified
encode_prompt in 54 papers data_generation/data_generation.py ran · our draft was wrong Apache-2.0 (permissive) · bde7f9e76c0f398c · report fixture: signature code_snippets:list/1/str, instructions:list/1/str, mode:str/0/str · returned str · level recorded 2026-09-22
post_process_1st_response data_generation/data_generation.py ran · our draft was wrong Apache-2.0 (permissive) · 0396d7986c9245e0 · report fixture: signature response:dict/0/?, example_ids:list/1/str · returned dict · level recorded 2026-09-22
generate_instruction_following_data data_generation/data_generation.py unverified Apache-2.0 (permissive) · 750d305ce32e0643 · report
num_tokens_from_messages in 14 papers data_generation/data_generation.py unverified Apache-2.0 (permissive) · da03aab82afbcc6b · report
one_call data_generation/data_generation.py unverified Apache-2.0 (permissive) · 0fd35c82026b1481 · report
imoneoi/openchat community (archive-listed) 3 samples, 1 ran (0 instrument-failed) · 2 unverified
ModelDataConfig ochat/experimental/generate_dataset_old.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 1afe396e11afa85e · report fixture: returned ModelDataConfig · level recorded 2026-09-22
generate_dataset in 41 papers ochat/experimental/generate_dataset_old.py unverified Apache-2.0 (permissive) · a035d0e0d1e7dd67 · report
generate_split in 9 papers ochat/experimental/generate_dataset_old.py unverified Apache-2.0 (permissive) · a4fc9157b8751af4 · report
jondurbin/airoboros community (archive-listed) 4 samples, 3 ran (0 instrument-failed) · 1 unverified
average_pool in 8 papers airoboros/embeddings.py ran fingerprinted Apache-2.0 (permissive) · 4c1e8ffec89c5351 · report fixture: signature last_hidden_states:Tensor/3/float, attention_mask:Tensor/2/int · returned Tensor · level recorded 2026-09-25
calculate_embeddings airoboros/embeddings.py ran Apache-2.0 (permissive) · d2f185ffba2db890 · report fixture: signature input_text:str/0/str, model:MockEmbeddingModel/0/?, tokenizer:callable/0/? · returned ndarray · level recorded 2026-09-25
calculate_fragment_embeddings airoboros/embeddings.py ran Apache-2.0 (permissive) · deed53aece8a579f · report fixture: signature model:MockEmbeddingModel/0/?, fragment:str/0/str · returned list · level recorded 2026-09-25
generate_prompt in 110 papers airoboros/instructors/contextual.py unverified Apache-2.0 (permissive) · 3df9ac239aea5147 · 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. 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. 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.

4 assemblies on this paper: 4 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

4 sibling assemblies for SELF-INSTRUCT (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for SELF-INSTRUCT (graph method: SELF-INSTRUCT). 3 bricks; 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) · 2 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 4 siblings

  1. generate_dataset entry · level 0 · unverified · a035d0e0d1e7dd67
  2. generate_split helper · level 0 · unverified · a4fc9157b8751af4
  3. ModelDataConfig helper · level 1 · ran · 1afe396e11afa85e

harvested Assembly for SELF-INSTRUCT (graph method: SELF-INSTRUCT). 6 bricks; individually on a fixture: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 4 siblings

  1. generate_instruction_following_data entry · level 0 · unverified · 750d305ce32e0643
  2. one_call sampler · level 0 · unverified · 0fd35c82026b1481
  3. encode_prompt helper · level 1 · ran · our draft was wrong · bde7f9e76c0f398c
  4. num_tokens_from_messages helper · level 0 · unverified · da03aab82afbcc6b
  5. post_process_1st_response data_transform · level 1 · ran · our draft was wrong · 0396d7986c9245e0
  6. find_word_in_string helper · level 1 · ran · our draft was wrong · 8ca6a5f9f77053c2

harvested Assembly for SELF-INSTRUCT (graph method: SELF-INSTRUCT). 3 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) · 1 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 4 siblings

  1. instruct_prompt entry · level 1 · ran · f56f83bd10a33749
  2. printf helper · level 0 · unverified · f53b42abf295354b
  3. prompt helper · level 1 · ran · 1171a8f3cf445159

harvested Assembly for SELF-INSTRUCT (graph method: SELF-INSTRUCT). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 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 4 siblings

  1. encode_prompt entry · level 1 · ran · our draft was wrong · e0456866b386ce58

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="2212.10560") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Instruction FollowingLanguage Modelling

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