Papers › Fine Tuning vs. Retrieval Augmented Generation for Less Popular Knowledge

Fine Tuning vs. Retrieval Augmented Generation for Less Popular Knowledge

3 Mar 2024arXiv:2403.01432archive 2025-07-28

Heydar Soudani, Evangelos Kanoulas, Faegheh Hasibi

Language Models (LMs) memorize a vast amount of factual knowledge, exhibiting strong performance across diverse tasks and domains. However, it has been observed that the performance diminishes when dealing with less-popular or low-frequency concepts and entities, for example in domain specific applications. The two prominent approaches to enhance the performance of LMs on low-frequent topics are: Retrieval Augmented Generation (RAG) and fine-tuning (FT) over synthetic data. This paper explores and evaluates the impact of RAG and FT on customizing LMs in handling low-frequency entities on question answering tasks. We conduct extensive experiments on twelve LMs of varying size and type and different fine tuning, data augmentation, and retrieval models. Our findings indicate that while FT boosts the performance across entities of varying popularity, RAG surpasses FT by a large margin particularly for least popular factual knowledge. Additionally, the success of both RAG and FT approaches is amplified by improving retrieval and data augmentation techniques. Fine tuning, while beneficial for small LMs, requires extensive resources. To address this issue, we propose the new Stimulus RAG approach that surpasses the effectiveness of fine tuning based approaches, thereby eliminating the need for the costly data augmentation and fine tuning step for enriching LMs with less popular factual knowledge. The code is available at https://github.com/informagi/RAGvsFT.

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 19 code samples harvested from 1 repository linked to this paper: 15 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 10 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 10 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 19 samples from 1 repository, 15 ran (5 with no instrument failure; 10 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.

heydarsoudani/ragvsft officialmentioned in paperpytorchSyntology: 19 samples, 15 ran (10 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

19 samples harvested: 15 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 10 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.

1ran · honoured contract
10ran · our draft was wrong
4ran
4unverified

Licence: 19 of the 19 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 heydarsoudani/ragvsft. “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.

heydarsoudani/ragvsft official repository 19 samples, 15 ran (10 instrument-failed) · 4 unverified
calculate_relative_popularity component0_preprocessing/relative_popularity_bucketing.py ran · our draft was wrong no licence file found · pointer only · 18e768963a04b1c3 · report fixture: signature objects:list/1/dict · returned list · level recorded 2026-09-24
calculated_accuracy analysis/plot_results.py ran · honoured contract no licence file found · pointer only · e1fee8e3b0d495f4 · report fixture: signature objects:list/1/dict · returned float · level recorded 2026-09-24
convert_to_url_format component0_preprocessing/create_costomized_dataset.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 885a140ff8e2fe75 · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-24
extract_json_objects component0_preprocessing/create_costomized_dataset.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 9a0d33580478460b · report fixture: signature text:str/0/str · returned list · level recorded 2026-09-24
load_dataset_qa component2_AnsGen/finetuning/flant5_finetuning.py ran no licence file found · pointer only · a41e1c043c80020e · report fixture: signature tokenizer:callable/0/?, test_files:dict/0/? · returned DatasetDict · level recorded 2026-09-24
load_dataset_qa component2_AnsGen/finetuning/llm_finetuning.py ran · our draft was wrong no licence file found · pointer only · adb13beccd33442d · report fixture: returned DatasetDict · level recorded 2026-09-24
load_json_file in 35 papers component2_AnsGen/finetuning/flant5_finetuning.py ran · our draft was wrong no licence file found · pointer only · ae696033188ab5b7 · report fixture: signature file_path:str/0/str · returned dict · level recorded 2026-09-24
load_json_file in 35 papers component2_AnsGen/finetuning/opt_finetuning.py ran · our draft was wrong no licence file found · pointer only · ee3add82700b2786 · report fixture: signature file_path:str/0/str · returned dict · level recorded 2026-09-24
load_json_file in 35 papers component3_highlighter/sentence_highlighter.py ran · our draft was wrong no licence file found · pointer only · 5de2b9b470eeba70 · report fixture: signature file_path:str/0/str · returned dict · level recorded 2026-09-24
load_json_files component2_AnsGen/finetuning/opt_finetuning.py ran no licence file found · pointer only · e9ad136bfabde8b4 · report fixture: signature relation_files:dict/0/?, split_name:str/0/str · returned dict · level recorded 2026-09-24
load_relations_data component2_AnsGen/finetuning/flant5_finetuning.py ran · our draft was wrong no licence file found · pointer only · 48b100358417247d · report fixture: signature args:Args/0/? · returned tuple · level recorded 2026-09-24
load_relations_data component3_highlighter/sentence_highlighter.py ran · our draft was wrong no licence file found · pointer only · 3137917c97622aa7 · report fixture: signature args:Args/0/? · returned tuple · level recorded 2026-09-24
split_by_relation component0_preprocessing/relative_popularity_bucketing.py ran · our draft was wrong no licence file found · pointer only · b099238f8b55282e · report fixture: signature input_file_path:str/0/str, output_file_path:str/0/str · returned dict · level recorded 2026-09-24
split_to_buckets analysis/plot_results.py ran no licence file found · pointer only · e353bb80690f02dd · report fixture: signature objects:list/1/dict, split_points:list/1/float · returned dict · level recorded 2026-09-24
split_to_buckets component1_retrieval/utils.py ran no licence file found · pointer only · b0f02c29c4153f5c · report fixture: signature objects:list/1/dict, split_points:list/1/float, args:Args/0/? · returned dict · level recorded 2026-09-24
get_wikipedia_title_from_wikidata component0_preprocessing/create_costomized_dataset.py unverified no licence file found · pointer only · 02f693941998e280 · report
main in 235 papers component1_retrieval/data_postprocessing.py unverified no licence file found · pointer only · d8772189b543b83f · report
sig_test analysis/plot_results.py unverified no licence file found · pointer only · ea3cc6f0ff764ddf · report
split_to_buckets component0_preprocessing/relative_popularity_bucketing.py unverified no licence file found · pointer only · 7018bfe9d2340be1 · report

Tasks

Data AugmentationQuestion AnsweringRAGRetrievalRetrieval-augmented Generation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AdamAttentionAttention DropoutBARTBERTBPEDense ConnectionsDropoutLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionRAGResidual ConnectionSoftmaxWeight DecayWordPiece

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