Seeing Like an AI: How LLMs Apply (and Misapply) Wikipedia Neutrality Norms

• URL: http://arxiv.org/abs/2407.04183v3
• Date: Sat, 14 Sep 2024 21:28:20 GMT
• Title: Seeing Like an AI: How LLMs Apply (and Misapply) Wikipedia Neutrality Norms
• Abstract summary: Large language models (LLMs) are trained on broad corpora and then used in communities with specialized norms. We evaluate their capacity to detect and correct biased Wikipedia edits according to Wikipedia's Neutral Point of View (NPOV) policy.
• Score: 11.01040577011255
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) are trained on broad corpora and then used in communities with specialized norms. Is providing LLMs with community rules enough for models to follow these norms? We evaluate LLMs' capacity to detect (Task 1) and correct (Task 2) biased Wikipedia edits according to Wikipedia's Neutral Point of View (NPOV) policy. LLMs struggled with bias detection, achieving only 64% accuracy on a balanced dataset. Models exhibited contrasting biases (some under- and others over-predicted bias), suggesting distinct priors about neutrality. LLMs performed better at generation, removing 79% of words removed by Wikipedia editors. However, LLMs made additional changes beyond Wikipedia editors' simpler neutralizations, resulting in high-recall but low-precision editing. Interestingly, crowdworkers rated AI rewrites as more neutral (70%) and fluent (61%) than Wikipedia-editor rewrites. Qualitative analysis found LLMs sometimes applied NPOV more comprehensively than Wikipedia editors but often made extraneous non-NPOV-related changes (such as grammar). LLMs may apply rules in ways that resonate with the public but diverge from community experts. While potentially effective for generation, LLMs may reduce editor agency and increase moderation workload (e.g., verifying additions). Even when rules are easy to articulate, having LLMs apply them like community members may still be difficult.

Related papers

• How LLMs Distort Our Written Language [11.98582470524476]
  Large language models (LLMs) are used by over a billion people globally, most often to assist with writing.<n>We demonstrate that LLMs not only alter the voice and tone of human writing, but also consistently alter the intended meaning.
  arXiv  Detail & Related papers  (2026-03-18T18:01:52Z)
• WINELL: Wikipedia Never-Ending Updating with LLM Agents [72.14647217009066]
  WiNELL is an agentic framework for continuously updating Wikipedia articles.<n>Our approach employs a multi-agent framework to aggregate online information, select new and important knowledge for a target entity in Wikipedia, and then generate precise edit suggestions for human review.
  arXiv  Detail & Related papers  (2025-07-30T07:51:42Z)
• DIF: A Framework for Benchmarking and Verifying Implicit Bias in LLMs [1.89915151018241]
  We argue that implicit bias in Large Language Models (LLMs) is not only an ethical, but also a technical issue.<n>We developed a method for calculating an easily interpretable benchmark, DIF (Demographic Implicit Fairness)
  arXiv  Detail & Related papers  (2025-05-15T06:53:37Z)
• Wikipedia in the Era of LLMs: Evolution and Risks [2.7204177341698266]
  We analyze the impact of Large Language Models (LLMs) on Wikipedia through existing data and using simulations to explore potential risks. Our findings and simulation results reveal that Wikipedia articles have been influenced by LLMs, with an impact of approximately 1%-2% in certain categories.
  arXiv  Detail & Related papers  (2025-03-04T18:58:13Z)
• IssueBench: Millions of Realistic Prompts for Measuring Issue Bias in LLM Writing Assistance [30.25793801015166]
  IssueBench is a set of 2.49m realistic prompts for measuring issue bias in large language models. We show that issue biases are common and persistent in state-of-the-art LLMs. All models align more with US Democrat than Republican voter opinion on a subset of issues.
  arXiv  Detail & Related papers  (2025-02-12T13:37:03Z)
• Mitigating Heterogeneous Token Overfitting in LLM Knowledge Editing [21.143790515287392]
  Large language models (LLMs) have achieved remarkable performance on various natural language tasks. They are trained on static corpora and their knowledge can become outdated quickly in the fast-changing world. This motivates the development of knowledge editing (KE) to update specific knowledge in LLMs without changing unrelated others or compromising their pre-trained capabilities.
  arXiv  Detail & Related papers  (2025-02-02T00:10:51Z)
• Learning to Rewrite: Generalized LLM-Generated Text Detection [19.9477991969521]
  Large language models (LLMs) can be abused at scale to create non-factual content and spread disinformation. We propose training an LLM to rewrite input text, producing minimal edits for LLM-generated content and more edits for human-written text. Our work suggests that LLM can effectively detect machine-generated text if they are trained properly.
  arXiv  Detail & Related papers  (2024-08-08T05:53:39Z)
• Can Editing LLMs Inject Harm? [122.83469484328465]
  We propose to reformulate knowledge editing as a new type of safety threat for Large Language Models. For the risk of misinformation injection, we first categorize it into commonsense misinformation injection and long-tail misinformation injection. For the risk of bias injection, we discover that not only can biased sentences be injected into LLMs with high effectiveness, but also one single biased sentence injection can cause a bias increase.
  arXiv  Detail & Related papers  (2024-07-29T17:58:06Z)
• WikiContradict: A Benchmark for Evaluating LLMs on Real-World Knowledge Conflicts from Wikipedia [59.96425443250666]
  Retrieval-augmented generation (RAG) has emerged as a promising solution to mitigate the limitations of large language models (LLMs) In this work, we conduct a comprehensive evaluation of LLM-generated answers to questions based on contradictory passages from Wikipedia. We benchmark a diverse range of both closed and open-source LLMs under different QA scenarios, including RAG with a single passage, and RAG with 2 contradictory passages.
  arXiv  Detail & Related papers  (2024-06-19T20:13:42Z)
• Editing Conceptual Knowledge for Large Language Models [65.38231526537476]
  This paper pioneers the investigation of editing conceptual knowledge for Large Language Models (LLMs) We construct a novel benchmark dataset ConceptEdit and establish a suite of new metrics for evaluation. experimental results reveal that, although existing editing methods can efficiently modify concept-level definition to some extent, they also have the potential to distort the related instantial knowledge.
  arXiv  Detail & Related papers  (2024-03-10T16:57:10Z)
• Pride and Prejudice: LLM Amplifies Self-Bias in Self-Refinement [75.7148545929689]
  Large language models (LLMs) improve their performance through self-feedback on certain tasks while degrade on others. We formally define LLM's self-bias - the tendency to favor its own generation. We analyze six LLMs on translation, constrained text generation, and mathematical reasoning tasks.
  arXiv  Detail & Related papers  (2024-02-18T03:10:39Z)
• See the Unseen: Better Context-Consistent Knowledge-Editing by Noises [73.54237379082795]
  Knowledge-editing updates knowledge of large language models (LLMs) Existing works ignore this property and the editing lacks generalization. We empirically find that the effects of different contexts upon LLMs in recalling the same knowledge follow a Gaussian-like distribution.
  arXiv  Detail & Related papers  (2024-01-15T09:09:14Z)
• Enabling Large Language Models to Learn from Rules [99.16680531261987]
  We are inspired that humans can learn the new tasks or knowledge in another way by learning from rules. We propose rule distillation, which first uses the strong in-context abilities of LLMs to extract the knowledge from the textual rules. Our experiments show that making LLMs learn from rules by our method is much more efficient than example-based learning in both the sample size and generalization ability.
  arXiv  Detail & Related papers  (2023-11-15T11:42:41Z)
• Are Large Language Models Temporally Grounded? [38.481606493496514]
  We provide Large language models (LLMs) with textual narratives. We probe them with respect to their common-sense knowledge of the structure and duration of events. We evaluate state-of-the-art LLMs on three tasks reflecting these abilities.
  arXiv  Detail & Related papers  (2023-11-14T18:57:15Z)
• Large Language Models Help Humans Verify Truthfulness -- Except When They Are Convincingly Wrong [35.64962031447787]
  Large Language Models (LLMs) are increasingly used for accessing information on the web. Our experiments with 80 crowdworkers compare language models with search engines (information retrieval systems) at facilitating fact-checking. Users reading LLM explanations are significantly more efficient than those using search engines while achieving similar accuracy.
  arXiv  Detail & Related papers  (2023-10-19T08:09:58Z)
• Gender bias and stereotypes in Large Language Models [0.6882042556551611]
  This paper investigates Large Language Models' behavior with respect to gender stereotypes. We use a simple paradigm to test the presence of gender bias, building on but differing from WinoBias. Our contributions in this paper are as follows: (a) LLMs are 3-6 times more likely to choose an occupation that stereotypically aligns with a person's gender; (b) these choices align with people's perceptions better than with the ground truth as reflected in official job statistics; (d) LLMs ignore crucial ambiguities in sentence structure 95% of the time in our study items, but when explicitly prompted, they recognize
  arXiv  Detail & Related papers  (2023-08-28T22:32:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.