Papers › Generative Parameter-Efficient Fine-Tuning

Generative Parameter-Efficient Fine-Tuning

1 Dec 2023arXiv:2312.00700archive 2025-07-28

Chinmay Savadikar, Xi Song, Tianfu Wu

We present Generative Parameter-Efficient Fine-Tuning (GIFT) for adapting pretrained Transformer backbones on downstream tasks. GIFT learns to generate the fine-tuned weights for a layer directly from its pretrained weights. The GIFT network is parameterized in a minimally-simple way by two linear layers (without bias terms), and is shared by different pretrained layers selected for fine-tuning (e.g., the Query layers), which result in significantly fewer trainable parameters compared to the layer-specific methods like Low-Rank Adapter (LoRA). We also show this formulation bridges parameter-efficient fine-tuning and representation fine-tuning. We perform comprehensive experiments on natural language tasks (commonsense and arithmetic reasoning, instruction tuning, and sequence classification) and computer vision tasks (fine-grained classification). We obtain the best performance and parameter efficiency among baselines on commonsense and arithmetic reasoning, and instruction following using the Llama family of models and on visual recognition benchmarks using Vision Transformers. Notably, compared to LoRA, we obtain 5.7% absolute increase in average accuracy with 14 times reduction of parameters on Commonsense170k using Llama-3 (8B), and 5.4% absolute increase in the win rate with 4 times reduction of parameters using Llama-2 (7B) during instruction tuning. Our GIFT also obtains a slightly higher win rate on instruction tuning than GPT 3.5 (Turbo 1106).

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 16 code samples harvested from 1 repository linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 2 honoured, 0 violated, 4 with no contract checked; 6 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 16 samples from 1 repository, 12 ran (6 with no instrument failure; 6 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 between 2026-09-24 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

savadikarc/gift officialmentioned in papermentioned on GitHubpytorchMITSyntology: 16 samples, 12 ran (6 instrument-failed) · 4 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

16 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 2 honoured, 0 violated, 4 with no contract checked; 6 where Syntology's instrument failed) · 4 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.

2ran · honoured contract
6ran · our draft was wrong
4ran
4unverified

Licence: 10 of the 16 samples are pointer only, meaning Syntology does not serve that copy's text. For 10 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from savadikarc/gift. “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.

savadikarc/gift official repository 16 samples, 12 ran (6 instrument-failed) · 4 unverified
SeqToSeqEncode language_modeling/math_code_instruct/utils.py ran MIT (permissive) here; the stored copy is pointer only · 45db30ac52f3cd6d · report fixture: signature example:dict/0/?, tokenizer:callable/0/?, max_length:int/0/int, ignore_masked_token:bool/0/bool · returned dict · level recorded 2026-09-26
cache_to_disk language_modeling/math_code_instruct/data.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 1d0db22ae067b53c · report fixture: signature root_datadir:str/0/str · returned function · level recorded 2026-09-26
causalLMEncode language_modeling/math_code_instruct/utils.py ran MIT (permissive) here; the stored copy is pointer only · 38de4191703bb96a · report fixture: signature example:dict/0/?, tokenizer:callable/0/?, max_length:int/0/int, ignore_masked_token:bool/0/bool · returned dict · level recorded 2026-09-26
extract_num in 5 papers language_modeling/math_code_instruct/eval_gsm8k.py ran · honoured contract fingerprinted MIT (permissive) · 9221df7ba44d3b36 · report fixture: signature text:str/0/str · returned int · level recorded 2026-09-24
find_all_linear_modules in 5 papers language_modeling/math_code_instruct/run_exp.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 65f3cc97d3e5034f · report fixture: signature model:callable/0/? · returned list · level recorded 2026-09-26
find_hidden_state_size language_modeling/math_code_instruct/run_exp.py ran · honoured contract MIT (permissive) · 4ce9d0bdc3c04522 · report fixture: signature model:callable/0/? · returned int · level recorded 2026-09-24
format_example in 43 papers language_modeling/math_code_instruct/eval_mmlu.py ran MIT (permissive) · dec15cc75848a98e · report fixture: signature df:DataFrame/0/?, idx:int/0/int, include_answer:bool/0/bool · returned str · level recorded 2026-09-24
format_subject in 26 papers language_modeling/math_code_instruct/eval_mmlu.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · 793a8b6496415457 · report fixture: signature subject:str/0/str · returned str · level recorded 2026-09-24
gen_prompt in 35 papers language_modeling/math_code_instruct/eval_mmlu.py ran · our draft was wrong MIT (permissive) · 04fd5045fce0c33a · report fixture: signature train_df:DataFrame/0/?, subject:str/0/str, k:int/0/int · returned str · level recorded 2026-09-24
get_forward_hook language_modeling/math_code_instruct/logTrainer.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · fd33ad9ae547c836 · report fixture: signature name:str/0/str · returned function · level recorded 2026-09-26
get_module in 37 papers language_modeling/math_code_instruct/lora_plus.py ran MIT (permissive) here; the stored copy is pointer only · ef1ba16faaacbb49 · report fixture: signature name:str/0/str, opt_model:callable/0/? · returned Linear · level recorded 2026-09-26
post_process in 56 papers language_modeling/math_code_instruct/eval_humaneval.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · 48720117c27620bc · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-26
create_loraplus_optimizer language_modeling/math_code_instruct/lora_plus.py unverified MIT (permissive) · 0921ef036f2219f1 · report
load_flan_v2 language_modeling/math_code_instruct/data.py unverified MIT (permissive) here; the stored copy is pointer only · 1b37d30c5eaded49 · report
load_meta_math language_modeling/math_code_instruct/data.py unverified MIT (permissive) here; the stored copy is pointer only · 963559e86ce91fd8 · report
reinit_lora language_modeling/math_code_instruct/run_exp.py unverified MIT (permissive) · e70a2129bcf0f57c · report

Tasks

Arithmetic ReasoningFine-Grained Image ClassificationInstruction FollowingSemantic Segmentationparameter-efficient fine-tuning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAdapterAttentionAttention DropoutBPECosine AnnealingDense ConnectionsDiscriminative Fine-TuningDropoutGPTLLaMALabel SmoothingLayer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformerVision TransformerWeight 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