WinoBias
Models resolve pronouns linked to occupations in type 1 and type 2 sentences. Accuracy gaps show gender bias in coreference.
Gemma reports give type 1 and type 2 scores separately.
- Made by
- UCLA (Zhao et al.)
- Released
- April 2018
- Kind
- Bias, fairness and politics
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Last in a system card
- July 2024
- Last model added
- July 2024
In the authors’ words
We introduce a new benchmark, WinoBias, for coreference resolution focused on gender bias. Our corpus contains Winograd-schema style sentences with entities corresponding to people referred by their occupation (e.g. the nurse, the doctor, the carpenter). We demonstrate that a rule-based, a feature-rich, and a neural coreference system all link gendered pronouns to pro-stereotypical entities with higher accuracy than anti-stereotypical entities, by an average difference of 21.1 in F1 score. Finally, we demonstrate a data-augmentation approach that, in combination with existing word-embedding debiasing techniques, removes the bias demonstrated by these systems in WinoBias without significantly affecting their performance on existing coreference benchmark datasets. Our dataset and code are available at http://winobias.org.
No scores here yet
No lab, leaderboard or Epoch AI run has reported a score for WinoBias yet.
Labs report it only in system cards
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.
4 system cards report it, from 1 lab
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Jul 31, 2024 |  Gemma 2 9B technical report | Winobias, top-1, IT, two subsets with no overall score: 1_2 78.09, 2_2 95.32 (Table 18) | Bias | |
| Jul 31, 2024 | Gemma 2 27B technical report | Winobias, top-1, IT, two subsets with no overall score: 1_2 81.94, 2_2 97.22 (Table 18) | Bias | |
| Mar 13, 2024 | Gemma 7B technical report | Winobias, IT, two subsets with no overall score: 1_2 59.09, 2_2 92.23 (Table 8) | Bias | |
| Mar 13, 2024 | Gemma 2B technical report | Winobias, IT, two subsets with no overall score: 1_2 56.12, 2_2 91.1 (Table 8) | Bias |
More in Safety and alignment
- FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
- StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is.
Inkling 98.6% - Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. Gemini 4 Carbon 0%
- DistressBench In use Models reply as a chatbot to 718 simulated users in psychological crisis, in single-turn and multi-turn chats. Three judge models score replies against clinician-written rubrics.
Muse Spark 1.2 88.4% - Lakera B3 (Backbone Breaker Benchmark) In use Crowdsourced prompt-injection attacks hit backbone LLMs inside agent threat snapshots at three defense levels. The score is attack resistance, 100 minus the board's vulnerability percentage. Claude Opus 4.5 89%
- IHEval In use Models face conflicts between system, user and tool instructions. Reports a composite score for following the higher-priority instruction. GPT-4o 70%