Papers › Generalized test utilities for long-tail performance in extreme multi-label classification

Generalized test utilities for long-tail performance in extreme multi-label classification

9 Nov 2023NeurIPS 2023 11arXiv:2311.05081archive 2025-07-28

Erik Schultheis, Marek Wydmuch, Wojciech Kotłowski, Rohit Babbar, Krzysztof Dembczyński

Extreme multi-label classification (XMLC) is the task of selecting a small subset of relevant labels from a very large set of possible labels. As such, it is characterized by long-tail labels, i.e., most labels have very few positive instances. With standard performance measures such as precision@k, a classifier can ignore tail labels and still report good performance. However, it is often argued that correct predictions in the tail are more "interesting" or "rewarding," but the community has not yet settled on a metric capturing this intuitive concept. The existing propensity-scored metrics fall short on this goal by confounding the problems of long-tail and missing labels. In this paper, we analyze generalized metrics budgeted "at k" as an alternative solution. To tackle the challenging problem of optimizing these metrics, we formulate it in the expected test utility (ETU) framework, which aims to optimize the expected performance on a fixed test set. We derive optimal prediction rules and construct computationally efficient approximations with provable regret guarantees and robustness against model misspecification. Our algorithm, based on block coordinate ascent, scales effortlessly to XMLC problems and obtains promising results in terms of long-tail performance.

PaperPDFConference PDFCodeCode 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 expected test utility (ETU) framework, generated from the paper text (not the authors' code), at level 2 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)” (quoted from Syntology's server card). No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='expected test utility (ETU) framework') (how to connect); the free check have('2311.05081') needs no token.

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

Code

No sample from the archive's official repository (mwydmuch/xcolumns) is linked to this paper; the 40 samples below come from 1 community repository. Syntology harvested code from mwydmuch/xcolumns for another paper.

mwydmuch/xcolumns: Syntology's graph links these samples to Consistent algorithms for multi-label classification with macro-at-k metrics (arXiv:2401.16594; that paper's own run record: 0 ran · 3 unverified); nothing checks that they implement this paper's method.

Syntology 40 code samples harvested from 1 repository linked to this paper: 17 ran (of which 1 constructed an object rather than computing a result; 5 with no instrument failure: 3 honoured, 1 violated, 1 with no contract checked; 12 where Syntology's instrument failed) · 23 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; 5 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 40 samples from 1 repository, 17 ran (5 with no instrument failure; 12 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 1 of its repositories that Syntology harvested for other papers:

Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

mwydmuch/xcolumns officialmentioned in papermentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Consistent algorithms for multi-label classification with macro-at-k metrics (arXiv:2401.16594; that paper's own run record: 0 ran · 3 unverified); nothing checks that they implement this paper's method. report
mwydmuch/macro-measures-in-xmlc mentioned on GitHubpytorchSyntology: 40 samples, 17 ran (12 instrument-failed) · 23 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.

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

40 samples harvested: 17 ran (of which 1 constructed an object rather than computing a result; 5 with no instrument failure: 3 honoured, 1 violated, 1 with no contract checked; 12 where Syntology's instrument failed) · 23 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.

3ran · honoured contract
1ran · violated contract
7ran · our draft was wrong
5ran; the contract check could not run
1ran
23unverified

Licence: 0 of the 40 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 mwydmuch/macro-measures-in-xmlc. “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”. For 7 samples that ran, the graph records a signature this page does not show: an argument's recorded type or element kind is a printout of an object in Syntology's harness (such as a mock object, a class or a pandas printout), not the type of an input, so it does not describe the call's shape. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

mwydmuch/macro-measures-in-xmlc community (archive-listed) 40 samples, 17 ran (12 instrument-failed) · 23 unverified
ConfusionMatrix xcolumns/frank_wolfe.py ran MIT (permissive) · f8dd1f41eb10e704 · report fixture: returned ConfusionMatrix · level recorded 2026-09-22
_calculate_conf_mat_entry xcolumns/frank_wolfe.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant MIT (permissive) · aadf803574231500 · report fixture: returned ndarray · level recorded 2026-09-22
_calculate_fn_dense xcolumns/frank_wolfe.py ran · our draft was wrong MIT (permissive) · d3c5b814277a9e26 · report fixture: returned ndarray · level recorded 2026-09-22
_calculate_fp_dense xcolumns/frank_wolfe.py ran · our draft was wrong fingerprinted MIT (permissive) · 980f7510a1e06c60 · report fixture: signature y_true:ndarray/2/float, y_pred:ndarray/2/float, axis:int/0/int, dtype:NoneType/0/? · returned ndarray · level recorded 2026-09-22
_calculate_tp_dense xcolumns/frank_wolfe.py ran · our draft was wrong fingerprinted MIT (permissive) · e0390de4ace2fe3b · report fixture: signature y_true:ndarray/2/float, y_pred:ndarray/2/float, axis:int/0/int, dtype:NoneType/0/? · returned ndarray · level recorded 2026-09-22
_metric_func_with_gradient_autograd xcolumns/frank_wolfe.py ran · honoured contract MIT (permissive) · 4d3d74b62ac8f5e4 · report fixture: signature metric_func:callable/0/?, tp:ndarray/0/float, fp:ndarray/0/float, fn:ndarray/0/float, tn:ndarray/0/float · returned tuple · level recorded 2026-09-22
_predict_using_randomized_weighted_classifier_np xcolumns/frank_wolfe.py ran · our draft was wrong MIT (permissive) · a3268e6e215d1c99 · report fixture: returned ndarray · level recorded 2026-09-22
_predict_weighted_per_instance_dense xcolumns/frank_wolfe.py ran · our draft was wrong MIT (permissive) · 35558fd29c65e169 · report fixture: returned ndarray · level recorded 2026-09-22
calculate_confusion_matrix in 5 papers xcolumns/frank_wolfe.py ran; the contract check could not run (RAISES) MIT (permissive) · f77c581447ce8785 · report fixture: signature y_true:ndarray/2/float, y_pred:ndarray/2/float, normalize:bool/0/bool, axis:int/0/int, dtype:NoneType/0/? · returned ConfusionMatrix · level recorded 2026-09-22
calculate_fn xcolumns/frank_wolfe.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant MIT (permissive) · a771cf410b5301de · report fixture: returned ndarray · level recorded 2026-09-22
calculate_fp xcolumns/frank_wolfe.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant MIT (permissive) · 67c1a50ecde50f09 · report fixture: returned ndarray · level recorded 2026-09-22
calculate_tp xcolumns/frank_wolfe.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant MIT (permissive) · 967d26afea1749f4 · report fixture: returned ndarray · level recorded 2026-09-22
str_bool_true xcolumns/frank_wolfe.py ran · violated contract MIT (permissive) · 3c9f3cc506e4d8ed · report fixture: signature value:str/0/str · returned bool · level recorded 2026-09-22
ternary_search xcolumns/frank_wolfe.py ran · honoured contract MIT (permissive) · 045501f3a185770d · report fixture: signature low:float/0/float, high:float/0/float, eps:float/0/float, func:callable/0/? · returned tuple · level recorded 2026-09-22
uniform_search xcolumns/frank_wolfe.py ran · honoured contract MIT (permissive) · ad4673e83380d7b5 · report fixture: signature low:float/0/float, high:float/0/float, step:float/0/float, func:callable/0/? · returned tuple · level recorded 2026-09-22
unpack_csr_matrix xcolumns/frank_wolfe.py ran · our draft was wrong MIT (permissive) · dec83cfff3139a49 · report fixture: signature matrix:csr_matrix/2/float · returned tuple · level recorded 2026-09-22
zeros_like xcolumns/frank_wolfe.py ran · our draft was wrong fingerprinted MIT (permissive) · 8be2535e47e2319f · report fixture: signature a:ndarray/2/float, shape:NoneType/0/?, dtype:NoneType/0/? · returned ndarray · level recorded 2026-09-22
RandomizedWeightedClassifier xcolumns/frank_wolfe.py unverified MIT (permissive) · eae7134a9338e69c · report
_calculate_fn_csr xcolumns/frank_wolfe.py unverified MIT (permissive) · 6a437326b2853c18 · report
_calculate_fp_csr xcolumns/frank_wolfe.py unverified MIT (permissive) · 46d480d4afbb7486 · report
_calculate_tp_csr xcolumns/frank_wolfe.py unverified MIT (permissive) · 1886195b5bef94c2 · report
_find_best_alpha xcolumns/frank_wolfe.py unverified MIT (permissive) · be8bdd0c31d9f25e · report
_predict_using_randomized_weighted_classifier_csr xcolumns/frank_wolfe.py unverified MIT (permissive) · 7edcfacb7a6dd49b · report
_predict_weighted_per_instance_csr xcolumns/frank_wolfe.py unverified MIT (permissive) · e4f75aae98b15d2b · report
find_classifier_using_fw xcolumns/frank_wolfe.py unverified MIT (permissive) · a5c7600a012167a3 · report
log_info in 5 papers xcolumns/frank_wolfe.py unverified MIT (permissive) · 504deb2e28f93cbc · report
numba_argtopk_csr xcolumns/frank_wolfe.py unverified MIT (permissive) · b8f054c37af5d33d · report
numba_calculate_sum_csr_mat_mul_mat xcolumns/frank_wolfe.py unverified MIT (permissive) · 4eb08eb9aeeb6206 · report
numba_calculate_sum_csr_mat_mul_ones_minus_mat xcolumns/frank_wolfe.py unverified MIT (permissive) · 82b05bba7f433cc3 · report
numba_csr_vec_mul_ones_minus_vec xcolumns/frank_wolfe.py unverified MIT (permissive) · 687f421957a9802f · report
numba_csr_vec_mul_vec xcolumns/frank_wolfe.py unverified MIT (permissive) · 5c74c11f6f018561 · report
numba_predict_weighted_per_instance_csr xcolumns/frank_wolfe.py unverified MIT (permissive) · a6ce3fbba16ea2fc · report
numba_predict_weighted_per_instance_csr_step xcolumns/frank_wolfe.py unverified MIT (permissive) · dd53a088e9b96bef · report
numba_resize xcolumns/frank_wolfe.py unverified MIT (permissive) · a80779c19991792d · report
numba_set_gains_csr xcolumns/frank_wolfe.py unverified MIT (permissive) · 6e08feeab5f41fbf · report
numba_topk_csr xcolumns/frank_wolfe.py unverified MIT (permissive) · 58c78794b61e760c · report
predict_using_randomized_weighted_classifier xcolumns/frank_wolfe.py unverified MIT (permissive) · d7fc8c0c37869d63 · report
predict_weighted_per_instance xcolumns/frank_wolfe.py unverified MIT (permissive) · b5a7308f9eca72df · report
unpack_csr_matrices xcolumns/frank_wolfe.py unverified MIT (permissive) · df2ecf91e2324681 · 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. 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. 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.

2 assemblies on this paper: 1 harvested, 1 generated.

generated Assembly for expected test utility (ETU) framework (graph method: expected test utility (ETU) framework), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer (level 2).

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

  1. compute_estimated_gain entry · level 2 · ran · 264179af1952353c0de888dc8775fd4dffaeb5633fc4c723486424962923f5ee

harvested Assembly for expected test utility (ETU) framework (graph method: expected test utility (ETU) framework). 40 bricks; individually on a fixture: 17 ran (of which 1 constructed an object rather than computing a result; 5 with no instrument failure: 3 honoured, 1 violated, 1 with no contract checked; 12 where Syntology's instrument failed) · 23 unverified. The assembly has not been run end-to-end.

  1. find_classifier_using_fw entry · level 0 · unverified · a5c7600a012167a3
  2. RandomizedWeightedClassifier model · level 0 · unverified · eae7134a9338e69c
  3. _find_best_alpha helper · level 0 · unverified · be8bdd0c31d9f25e
  4. _metric_func_with_gradient_autograd helper · level 1 · ran · honoured contract · 4d3d74b62ac8f5e4
  5. calculate_confusion_matrix helper · level 1 · ran; the contract check could not run (RAISES) · f77c581447ce8785
  6. log_info helper · level 0 · unverified · 504deb2e28f93cbc
  7. predict_weighted_per_instance helper · level 0 · unverified · b5a7308f9eca72df
  8. predict_using_randomized_weighted_classifier model · level 0 · unverified · d7fc8c0c37869d63
  9. ternary_search helper · level 1 · ran · honoured contract · 045501f3a185770d
  10. uniform_search helper · level 1 · ran · honoured contract · ad4673e83380d7b5
  11. ConfusionMatrix helper · level 1 · ran · f8dd1f41eb10e704
  12. calculate_fn helper · level 1 · ran; the contract check could not run (RAISES) · a771cf410b5301de
  13. calculate_fp helper · level 1 · ran; the contract check could not run (RAISES) · 67c1a50ecde50f09
  14. calculate_tp helper · level 1 · ran; the contract check could not run (RAISES) · 967d26afea1749f4
  15. log helper · level 0 · unverified · 01febb6c4f142e9d
  16. _predict_weighted_per_instance_csr helper · level 0 · unverified · e4f75aae98b15d2b
  17. _predict_weighted_per_instance_dense helper · level 1 · ran · our draft was wrong · 35558fd29c65e169
  18. _predict_using_randomized_weighted_classifier_csr helper · level 0 · unverified · 7edcfacb7a6dd49b
  19. _predict_using_randomized_weighted_classifier_np helper · level 1 · ran · our draft was wrong · a3268e6e215d1c99
  20. _calculate_conf_mat_entry helper · level 1 · ran; the contract check could not run (RAISES) · aadf803574231500
  21. _calculate_fn_csr helper · level 0 · unverified · 6a437326b2853c18
  22. _calculate_fn_dense helper · level 1 · ran · our draft was wrong · d3c5b814277a9e26
  23. _calculate_fp_csr helper · level 0 · unverified · 46d480d4afbb7486
  24. _calculate_fp_dense helper · level 1 · ran · our draft was wrong · 980f7510a1e06c60
  25. _calculate_tp_csr helper · level 0 · unverified · 1886195b5bef94c2
  26. _calculate_tp_dense helper · level 1 · ran · our draft was wrong · e0390de4ace2fe3b
  27. numba_predict_weighted_per_instance_csr helper · level 0 · unverified · a6ce3fbba16ea2fc
  28. unpack_csr_matrix helper · level 1 · ran · our draft was wrong · dec83cfff3139a49
  29. zeros_like helper · level 1 · ran · our draft was wrong · 8be2535e47e2319f
  30. numba_predict_weighted_per_instance_csr_step helper · level 0 · unverified · dd53a088e9b96bef
  31. numba_calculate_sum_csr_mat_mul_ones_minus_mat helper · level 0 · unverified · 82b05bba7f433cc3
  32. unpack_csr_matrices helper · level 0 · unverified · df2ecf91e2324681
  33. numba_calculate_sum_csr_mat_mul_mat helper · level 0 · unverified · 4eb08eb9aeeb6206
  34. numba_argtopk_csr helper · level 0 · unverified · b8f054c37af5d33d
  35. numba_topk_csr helper · level 0 · unverified · 58c78794b61e760c
  36. numba_set_gains_csr helper · level 0 · unverified · 6e08feeab5f41fbf
  37. numba_csr_vec_mul_ones_minus_vec helper · level 0 · unverified · 687f421957a9802f
  38. numba_csr_vec_mul_vec helper · level 0 · unverified · 5c74c11f6f018561
  39. numba_resize helper · level 0 · unverified · a80779c19991792d
  40. str_bool_true helper · level 1 · ran · violated contract · 3c9f3cc506e4d8ed

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="2311.05081") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Extreme Multi-Label ClassificationMUlTI-LABEL-ClASSIFICATIONMissing LabelsMulti-Label Classification

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

SET

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