SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2411.00680 · NeurIPS · 2024

Zipfian Whitening

Sho Yokoi, Han Bao, Hiroto Kurita, Hidetoshi Shimodaira

arXiv · PDF · Open in the Atlas

Code that ran

3 of 4 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 1 failure of our instrument; 1 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2411.00680.json · badge: /api/ran/2411.00680.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
cl-tohoku/zipfian-whitening canonical (classifier label)
classifier basis: census
3 ran · 1 no recorded run
FunctionStatusWhere it lives
cos_sim Ran
ran; our contract draft was wrong, not the code
cl-tohoku/zipfian-whitening/src/calc_isotropy_score.py
pointer only (licence: NONE) · get_code("de1b89dc518ebc26")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cl-tohoku/zipfian-whitening when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_diag_of_cov Ran
ran, violated the contract we drafted
cl-tohoku/zipfian-whitening/src/IsoScore.py
pointer only (licence: NONE); the stored copy get_code serves, from francois-meyer/cpt-geometry, is pointer only (licence: NONE) · get_code("b9edc5a66c170732")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cl-tohoku/zipfian-whitening when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

normalize_diagonal Ran
ran, violated the contract we drafted
cl-tohoku/zipfian-whitening/src/IsoScore.py
pointer only (licence: NONE) · get_code("8b4fb7a187d33e21")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cl-tohoku/zipfian-whitening when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

pca_normalization Not yet run
unverified (harvested, no recorded run)
cl-tohoku/zipfian-whitening/src/IsoScore.py
pointer only (licence: NONE) · get_code("5e0a2765ed1280c0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cl-tohoku/zipfian-whitening when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
get_diag_of_cov 2 other papers
2 later
2604.16576 (2026-04)
On the Robustness of LLM-Based Dense Retrievers: A Systematic Analysis of Generalizability and Stability
2608.23358 (2026-08)
The Geometry of Low-Resource Language Representations
normalize_diagonal 1 other paper
1 later
2604.16576 (2026-04)
On the Robustness of LLM-Based Dense Retrievers: A Systematic Analysis of Generalizability and Stability
pca_normalization 1 other paper
1 later
2604.16576 (2026-04)
On the Robustness of LLM-Based Dense Retrievers: A Systematic Analysis of Generalizability and Stability

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The word embedding space in neural models is skewed, and correcting this can improve task performance. We point out that most approaches for modeling, correcting, and measuring the symmetry of an embedding space implicitly assume that the word frequencies are uniform; in reality, word frequencies follow a highly non-uniform distribution, known as Zipf's law. Surprisingly, simply performing PCA whitening weighted by the empirical word frequency that follows Zipf's law significantly improves task performance, surpassing established baselines. From a theoretical perspective, both our approach and existing methods can be clearly categorized: word representations are distributed according to an exponential family with either uniform or Zipfian base measures. By adopting the latter approach, we can naturally emphasize informative low-frequency words in terms of their vector norm, which becomes evident from the information-geometric perspective [42], and in terms of the loss functions for imbalanced classification [36]. Additionally, our theory corroborates that popular natural language processing methods, such as skip-gram negative sampling [37], WhiteningBERT [26], and headless language models [23], work well just because their word embeddings encode the empirical word frequency into the underlying probabilistic model. https://github.com/cl-tohoku/zipfian-whitening 50 words sampled from each of types and tokens. Uniform sampling from types, corresponding to an unweighted mean, tends to select mostly rare words from the heavy tail. Sampling from tokens clearly captures a more natural representation of language as it typically appears in text. Table 1: The difference between type-based sampling and token-based sampling.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2411.00680")
get_code_for_paper("2411.00680")
have("2411.00680")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2411.00680.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2411.00680.svg)](https://syntology.ai/paper/2411.00680)

Connect an agent — have() is free.