Papers › Generalization v.s. Memorization: Tracing Language Models' Capabilities Back to...

Generalization v.s. Memorization: Tracing Language Models' Capabilities Back to Pretraining Data

20 Jul 2024arXiv:2407.14985archive 2025-07-28

Xinyi Wang, Antonis Antoniades, Yanai Elazar, Alfonso Amayuelas, Alon Albalak, Kexun Zhang, William Yang Wang

The impressive capabilities of large language models (LLMs) have sparked debate over whether these models genuinely generalize to unseen tasks or predominantly rely on memorizing vast amounts of pretraining data. To explore this issue, we introduce an extended concept of memorization, distributional memorization, which measures the correlation between the LLM output probabilities and the pretraining data frequency. To effectively capture task-specific pretraining data frequency, we propose a novel task-gram language model, which is built by counting the co-occurrence of semantically related n-gram pairs from task inputs and outputs in the pretraining corpus. Using the Pythia models trained on the Pile dataset, we evaluate four distinct tasks: machine translation, factual question answering, world knowledge understanding, and math reasoning. Our findings reveal varying levels of memorization, with the strongest effect observed in factual question answering. Furthermore, while model performance improves across all tasks as LLM size increases, only factual question answering shows an increase in memorization, whereas machine translation and reasoning tasks exhibit greater generalization, producing more novel outputs. This study demonstrates that memorization plays a larger role in simpler, knowledge-intensive tasks, while generalization is the key for harder, reasoning-based tasks, providing a scalable method for analyzing large pretraining corpora in greater depth. We also show the practical implications of our analysis through a novel prompt optimization algorithm.

PaperPDFCode 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.

Syntology holds a reference implementation of task-gram language model, generated from the paper text (not the authors' code), at level 3 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” (quoted from Syntology's server card). No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='task-gram language model') (how to connect); the free check have('2407.14985') needs no token.

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

Code

Syntology 15 code samples harvested from 1 repository linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 3 honoured, 1 violated, 0 with no contract checked; 7 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: found in paper text by Syntology: 15 samples from 1 repository, 11 ran (4 with no instrument failure; 7 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 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

EleutherAI/the-pile found in paper text by SyntologyMITSyntology: 15 samples, 11 ran (7 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

15 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 3 honoured, 1 violated, 0 with no contract checked; 7 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.

3ran · honoured contract
1ran · violated contract
7ran · our draft was wrong
4unverified

Licence: 0 of the 15 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 EleutherAI/the-pile. “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.

EleutherAI/the-pile found in paper text by Syntology 15 samples, 11 ran (7 instrument-failed) · 4 unverified
cont processing_scripts/fix_empty_lines.py ran · our draft was wrong fingerprinted MIT (permissive) · 128c815587eb6f9c · report fixture: signature x:str/0/str · returned str · level recorded 2026-09-24
despace processing_scripts/fix_dm_math.py ran · our draft was wrong fingerprinted MIT (permissive) · 400223165519340e · report fixture: signature x:str/0/str · returned str · level recorded 2026-09-24
fix in 7 papers processing_scripts/fix_dm_math.py ran · our draft was wrong MIT (permissive) · 102a3e6041b8359d · report fixture: signature x:bytes/0/? · returned bytes · level recorded 2026-09-24
id the_pile/utils.py ran · honoured contract fingerprinted MIT (permissive) · 79d53604e4138cba · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-24
mean in 84 papers processing_scripts/lang_len_analysis_pass2.py ran · violated contract fingerprinted MIT (permissive) · 0479fb5f08e12e14 · report fixture: signature x:list/1/float · returned float · level recorded 2026-09-24
readf processing_scripts/join.py ran · our draft was wrong MIT (permissive) · ea45516e7c59d783 · report fixture: signature f:str/0/str · returned list · level recorded 2026-09-24
rewrite_name processing_scripts/lang_len_analysis_pass2.py ran · our draft was wrong fingerprinted MIT (permissive) · 2ce3a6653bbabf47 · report fixture: signature n:str/0/str · returned str · level recorded 2026-09-24
stableorder the_pile/utils.py ran · our draft was wrong MIT (permissive) · 2246bc566da383d9 · report fixture: signature x:list/1/str · returned list · level recorded 2026-09-24
stddev processing_scripts/lang_len_analysis_pass2.py ran · our draft was wrong fingerprinted MIT (permissive) · 2f9c8f709caa01bc · report fixture: signature x:list/1/float · returned float · level recorded 2026-09-24
take in 8 papers the_pile/pile.py ran · honoured contract MIT (permissive) · 7ed0726edbb13bb5 · report fixture: signature n:int/0/int, iter:generator/0/? · returned list · level recorded 2026-09-22
utf8len the_pile/utils.py ran · honoured contract fingerprinted MIT (permissive) · 66211bd1a2260664 · report fixture: signature s:str/0/str · returned int · level recorded 2026-09-24
analyze in 11 papers processing_scripts/lang_len_analysis_pass1.py unverified MIT (permissive) · fc867bf5f0532c70 · report
json_parser the_pile/tfds_pile.py unverified MIT (permissive) · 5d4f4f3f6f2bb5f0 · report
language processing_scripts/lang_len_analysis_pass1.py unverified MIT (permissive) · c19e7511b7716d16 · report
lengths processing_scripts/lang_len_analysis_pass1.py unverified MIT (permissive) · 0ada8308c7e8d45e · 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. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.

1 assembly on this paper: 0 harvested, 1 generated.

generated Assembly for task-gram language model (graph method: task-gram language model), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer (level 3; 5 property tests passed on both).

label matches by construction: the code was generated for this method (not judged by a model)

  1. compute_ngram_memorization_score entry · level 3 · ran · honoured contract · fd109d5e91a25e8c13a9370d630bd330e64581939185d7fea88548ae8ac431eb

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="2407.14985") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Language ModellingMachine TranslationMathMemorizationMultiple-choiceQuestion AnsweringTranslationWorld Knowledge

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

FocusPythia

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