Papers › Towards Robustifying NLI Models Against Lexical Dataset Biases

Towards Robustifying NLI Models Against Lexical Dataset Biases

10 May 2020ACL 2020 6arXiv:2005.04732archive 2025-07-28

Xiang Zhou, Mohit Bansal

While deep learning models are making fast progress on the task of Natural Language Inference, recent studies have also shown that these models achieve high accuracy by exploiting several dataset biases, and without deep understanding of the language semantics. Using contradiction-word bias and word-overlapping bias as our two bias examples, this paper explores both data-level and model-level debiasing methods to robustify models against lexical dataset biases. First, we debias the dataset through data augmentation and enhancement, but show that the model bias cannot be fully removed via this method. Next, we also compare two ways of directly debiasing the model without knowing what the dataset biases are in advance. The first approach aims to remove the label bias at the embedding level. The second approach employs a bag-of-words sub-model to capture the features that are likely to exploit the bias and prevents the original model from learning these biased features by forcing orthogonality between these two sub-models. We performed evaluations on new balanced datasets extracted from the original MNLI dataset as well as the NLI stress tests, and show that the orthogonality approach is better at debiasing the model while maintaining competitive overall accuracy. Our code and data are available at: https://github.com/owenzx/LexicalDebias-ACL2020

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.

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

Code

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

By repository: official repository: 11 samples from 1 repository, 4 ran (0 with no instrument failure; 4 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 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

owenzx/LexicalDebias-ACL2020 officialmentioned in papermentioned on GitHubtfSyntology: 11 samples, 4 ran (4 instrument-failed) · 7 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

11 samples harvested: 4 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 4 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Every run on this page is a failure 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.

4ran · our draft was wrong
7unverified

Licence: 0 of the 11 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 owenzx/LexicalDebias-ACL2020. “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. 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.

owenzx/LexicalDebias-ACL2020 official repository 11 samples, 4 ran (4 instrument-failed) · 7 unverified
combine_last_two_dimensions models/hex.py ran · our draft was wrong fingerprinted MIT (permissive) · 4de7e27ebec50a84 · report fixture: signature x:EagerTensor/4/float · returned EagerTensor · level recorded 2026-09-22
length in 22 papers models/hex.py ran · our draft was wrong fingerprinted MIT (permissive) · bdffe2f0ce34eb31 · report fixture: signature sequence:EagerTensor/3/float · returned tuple · level recorded 2026-09-22
mask_logits in 27 papers
census: 7 implementations from 10 papers, 4 distinct outputs across 3 buckets, one shared input per bucket
models/hex.py ran · our draft was wrong fingerprinted MIT (permissive) · 1b1fb77042d2e192 · report fixture: signature inputs:EagerTensor/3/float, mask:EagerTensor/3/float, mask_value:float/0/float · returned EagerTensor · level recorded 2026-09-22
split_last_dimension models/hex.py ran · our draft was wrong fingerprinted MIT (permissive) · 625e74f352090a3a · report fixture: signature x:EagerTensor/3/int, n:int/0/int · returned EagerTensor · level recorded 2026-09-22
HEX models/hex.py unverified MIT (permissive) · 498eb1889a79901e · report
conv in 178 papers models/hex.py unverified MIT (permissive) · 413976febb345bb5 · report
dot_product_attention in 15 papers models/hex.py unverified MIT (permissive) · 0403fe04af127ad4 · report
layer_norm in 32 papers models/hex.py unverified MIT (permissive) · 11f9661c4bf99c8a · report
layer_norm_compute_python in 5 papers models/hex.py unverified MIT (permissive) · df5f1c26bcb94d35 · report
multihead_attention in 5 papers models/hex.py unverified MIT (permissive) · 0e1b744dc6c987e3 · report
simple_self_attention_block models/hex.py unverified MIT (permissive) · 4633aaac6d095124 · 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. 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.

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

harvested Assembly for orthogonality approach (graph method: orthogonality approach). 11 bricks; individually on a fixture: 4 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 4 where Syntology's instrument failed) · 7 unverified. The assembly has not been run end-to-end.

  1. HEX entry · level 0 · unverified · 498eb1889a79901e
  2. length helper · level 1 · ran · our draft was wrong · bdffe2f0ce34eb31
  3. simple_self_attention_block helper · level 0 · unverified · 4633aaac6d095124
  4. conv helper · level 0 · unverified · 413976febb345bb5
  5. layer_norm helper · level 0 · unverified · 11f9661c4bf99c8a
  6. multihead_attention helper · level 0 · unverified · 0e1b744dc6c987e3
  7. layer_norm_compute_python helper · level 0 · unverified · df5f1c26bcb94d35
  8. combine_last_two_dimensions helper · level 1 · ran · our draft was wrong · 4de7e27ebec50a84
  9. dot_product_attention helper · level 0 · unverified · 0403fe04af127ad4
  10. split_last_dimension helper · level 1 · ran · our draft was wrong · 625e74f352090a3a
  11. mask_logits helper · level 1 · ran · our draft was wrong · 1b1fb77042d2e192

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="2005.04732") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Data AugmentationNatural Language Inference

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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