Cited by 5Academic Writing and PublishingAcademic Publishing and Open AccessData Visualization and Analytics

What Makes a Good Poster? Evaluating #BetterPoster and Classic Formats at a Scientific Cancer Conference

Open access
Bentsen, Line, Østergaard, Daniella Elisabet·2026·Journal of Cancer Education

Abstract

Academic posters have evolved significantly, shifting from rare presentation opportunities to a common method of scientific dissemination. As speaking time and individualized attention for posters have diminished, new challenges have emerged, emphasizing the need for effective visual communication. This study aimed to evaluate the adoption and effectiveness of the #BetterPoster (BP) format compared to the classic format (C) and to determine whether the modified format contributed to an improved experience for both viewers and readers. At the national Danish Cancer Research Conference 2024, a total of 103 posters were categorized and assessed systematically by independent raters using four criteria: first impression, organization, poster format, and wordiness. Statistical analyses, including t -tests and Kendall’s W for interrater reliability, compared the formats, while Pearson correlations explored relationships between “first impression” and “organization,” “poster format,” or “wordiness.” Results indicated that the BP format achieved higher scores for all criteria, with significant differences in first impression ( p = 1.64e-06). However, both formats exhibited potential for high scores when designed mindfully, balancing text and visual elements. While the BP format improved audience engagement, excessive simplification or disorganization reduced its effectiveness. Classic posters, although often more text-heavy, could also be effective when structured carefully. The findings highlight the importance of coherent design and training in visual communication, suggesting that well-executed posters, regardless of format, can enhance academic engagement and knowledge dissemination. These insights could inform future guidelines for academic poster presentations, fostering more engaging and effective communication at conferences.

AI snapshot

Written from the paper’s full text.

TL;DR

This observational study assessed 103 posters at the Danish Cancer Research Conference 2024 and found that #BetterPoster-format posters scored higher than classic posters across first impression, organization, poster design, and wordiness. However, both formats could perform well when designed mindfully, and poor execution reduced effectiveness regardless of format.

Key findings

  • Among 103 posters, #BetterPoster posters had a mean first-impression score of 32.5 versus 25.7 for classic posters, with a significant difference (p = 1.64e-06).
  • Statistical testing showed significantly higher scores for the #BetterPoster format for all four assessed items: first impression, organization, poster design, and wordiness (table S3).
  • Interrater reliability was moderate to strong for first impression, poster design, and wordiness, but low for organization.
  • Both formats could reach high scores when used mindfully; lower ratings in #BetterPoster posters were linked to excessive simplification or disorganization, while lower ratings in classic posters were often linked to being overly text-heavy.

Methods

  • Observational study of 103 posters presented at the national Danish Cancer Research Conference 2024 in Odense; posters were categorized as #BetterPoster or classic format independently by two raters, with joint reassessment for disagreements.
  • Four items were assessed by two to four raters per poster: first impression (0–50 points), organization (1–4), poster design (1–4), and wordiness (1–4).
  • Analyses in R version 4.3.2 included Kendall’s W for interrater reliability, Shapiro-Wilk and Levene’s tests, t-tests or Wilcox tests for format comparisons, and Pearson correlations between first impression and organization, poster design, or wordiness.

Limitations

  • The authors state that evaluations were subjective, that raters including LB and DEØ had a preference for the #BetterPoster format, and that this may have biased scores; they also note no repeated evaluations over time, no systematic measurement of poster traffic, and that LB and DEØ had previously encouraged presenters to adopt the #BetterPoster format.
  • The authors state that additional forums of varying sizes would increase generalizability.
  • Evident from the text: initial consensus for categorizing poster format was only 57%, and interrater agreement for the organization item was low, making organization-related findings less reliable.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1007/s13187-025-02622-1

Cites

15 of 15 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

11 of 18 cited sources checked · 7 supported · 0 flagged · accuracy not yet measured

  • Findingown data
    § Poster Type and Scores · p. 4

    The linear regression slope between first impression and the other items was steeper for classic format than for #BetterPoster format.

    “However, the linear regression model was steeper for classic compared to better poster format.”
    Quote in the paperNot a substantial claim
  • Findingown data
    § Poster Type and Scores · p. 4

    #BetterPoster format had significantly higher scores than classic format across all four items (first impression, organization, poster design, wordiness).

    “Statistical testing for difference in scores for the two poster formats for all four items showed significantly higher scores for the better poster format (table S3).”
    • Also stated in: § Discussion · p. 4 · plain
      “In this study, based on the poster session at the national Danish Cancer Research Conference 2024, we found that the better poster format vs. the classic format scored significantly higher for first impression, organization, poster design, and wordiness.”
    Quote in the paperSubstantial claim
  • Findingown data
    § Discussion · p. 4

    Many classic format posters featured long blocks of text with conclusions at the bottom, requiring viewers to sift through information to grasp the key message.

    “Many posters with classic format featured long blocks of text with conclusions placed at the bottom, requiring viewers to sift through a great deal of information to grasp the key message.”
    Quote in the paperSubstantial claim
  • Findingown data
    § Poster Type and Scores · p. 4

    #BetterPoster format had significantly higher first impression scores than classic format (p = 1.64e-06).

    “A t -test showed significantly higher scores for #BetterPoster format ( p -value = 1.64e-06).”
    Quote in the paperSubstantial claim
  • Interpretationreasoning
    § Abstract · p. 1

    Excessive simplification or disorganization reduced the effectiveness of #BetterPoster posters.

    “While the BP format improved audience engagement, excessive simplification or disorganization reduced its effectiveness.”
    • Also stated in: § Discussion · p. 4 · hedged
      “Posters with better poster format could become disorganized, especially if random elements like excessive highlighting or ill-fitting images were used.”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. 22
  • May combine two claims. 16
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion · p. 4

    Khadka et al. emphasized that poster quality varied significantly, from poorly executed to exemplary, regardless of format.

    “Their study emphasized that the quality of posters varied significantly, ranging from poorly executed to exemplary, regardless of format [8].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [8] Khadka S, Holt K, Peeters MJ (2024) Academic conference posters: describing visual impression in pharmacy education [Internet]. Explor Res Clin Soc Pharm 13:100423. Available from: https://linkinghub.elsevier.com/retrieve/pii/S2667276624000180 source
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation
    § Items Assessed · p. 2

    The scales for organization, poster design, and wordiness were adapted from Khadka et al.

    “A detailed description of the scales for the items 'organization,' 'poster design,' and 'wordiness' were defined as seen in Table 1, inspired by Khadka et al. [8].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [8] Khadka S, Holt K, Peeters MJ (2024) Academic conference posters: describing visual impression in pharmacy education [Internet]. Explor Res Clin Soc Pharm 13:100423. Available from: https://linkinghub.elsevier.com/retrieve/pii/S2667276624000180 source
  • May combine two claims.

    Premisecitation
    § Discussion · p. 5

    Most researchers receive no design training and often little to no feedback when preparing visual materials for poster presentations.

    “Jambor et al. found that most researchers receive no design training and often receive little to no feedback when preparing their visual materials for poster presentations [20].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [20] Jambor HK. Insights on poster preparation practices in life sciences. Front Bioinforma [Internet]. 2023 Nov 1;3. Available from: https://www.frontiersin.org/articles/10.3389/fbinf.2023.1216139/full source
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion · p. 5

    Other researchers have emphasized the importance of first impressions at academic conferences.

    “Other researchers have emphasized the importance of first impressions at academic conferences.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [7] Smith PEM, Fuller G, Dunstan F. Scoring posters at scientific meetings: first impressions count. J R Soc Med [Internet]. 2004 Jul 23;97(7):340–1. Available from: http://journals.sagepub.com/doi/10.1177/014107680409700714 source
    • cites [13] Morrison M, Merlo K, Woessner Z (2020) How to boost the impact of scientific conferences [Internet]. Cell 182(5):1067–71. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0092867420309375 . Accessed Apr 2024 source

      The section “Focusing on User Goals Can Help Conferences Invent All-New Formats” lists meeting new people as a conference goal, but does not discuss the importance of first impressions. Searches for “first impressions” and for impressions when meeting or networking with people found no passage addressing that point. (LLM, conf=0.72)

      Experimental — accuracy not yet measured
  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisecitation
    § Discussion · p. 5

    Limiting the focus in a poster to two or three key findings is crucial for an effective research poster.

    “However, limiting the focus in the poster to two or three key findings is crucial for an effective research poster [16].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [16] Miller JE (2007) Preparing and presenting effective research posters [Internet]. Health Serv Res 20;42(1p1):311–28. Available from: https://onlinelibrary.wiley.com/doi/10.1111/j.1475-6773.2006.00588.x . Accessed Oct 2024 source
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation
    § Discussion · p. 4

    Khadka et al. (2024) found that contemporary poster formats such as the better poster format appeared more helpful in engaging the audience but were not without flaws.

    “Similar to our results, Khadka et al. (2024) found that while contemporary poster formats, such as the better poster format, appeared more helpful in engaging the audience, they were not without flaws.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [8] Khadka S, Holt K, Peeters MJ (2024) Academic conference posters: describing visual impression in pharmacy education [Internet]. Explor Res Clin Soc Pharm 13:100423. Available from: https://linkinghub.elsevier.com/retrieve/pii/S2667276624000180 source

      The Results report that contemporary posters received significantly higher visual-impression scores, and the Discussion says some contemporary posters nevertheless turned out badly. The study assessed visual impression at one pharmacy education conference, not audience engagement directly, so that part of the paraphrase goes beyond its findings. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation
    § Discussion · p. 5

    Only a small proportion of conference attendees take the time to read posters in detail or ask questions to presenters.

    “Another study showed that only a small proportion of conference attendees take the time to read posters in detail or ask questions to the presenters [14].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [14] Goodhand J, Giles C, Wahed M, Irving P, Langmead L, Rampton D (2011) Poster presentations at medical conferences: an effective way of disseminating research? Clin Med (Northfield Il) 11(2):138–41. Available from: https://linkinghub.elsevier.com/retrieve/pii/S1470211824053624 source
  • May combine two claims.

    Interpretationcitation
    § Discussion · p. 5

    Posters with a strong first impression are more likely to capture attention and facilitate further engagement.

    “This reinforces our finding that posters with a strong first impression, such as those mindfully using the better poster or classic format, are more likely to capture attention and facilitate further engagement.”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [7] Smith PEM, Fuller G, Dunstan F. Scoring posters at scientific meetings: first impressions count. J R Soc Med [Internet]. 2004 Jul 23;97(7):340–1. Available from: http://journals.sagepub.com/doi/10.1177/014107680409700714 source

      The Comment says an unattractive poster risked being overlooked on first impression and that specialists could use quick scores to select posters for more detailed assessment. This supports the idea that first impressions affect which posters receive further attention, but the paper did not measure audience engagement. The ‘better poster’ and classic formats are presented as the citing authors’ examples, not findings attributed to this source. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [13] Morrison M, Merlo K, Woessner Z (2020) How to boost the impact of scientific conferences [Internet]. Cell 182(5):1067–71. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0092867420309375 . Accessed Apr 2024 source

      The 'Posters Can Have a Second Life on Figshare' section discusses poster visitor counts and online reach, not first impressions or poster formats. The 'Social Media Posts Can Teach, Not Just Tantalize' section suggests that informative social-media posts may prompt click-throughs, but our copy does not address whether a strong first impression from better-poster or classic-format posters captures attention and leads to further engagement. (LLM, conf=0.74)

      Experimental — accuracy not yet measured
    • cites [14] Goodhand J, Giles C, Wahed M, Irving P, Langmead L, Rampton D (2011) Poster presentations at medical conferences: an effective way of disseminating research? Clin Med (Northfield Il) 11(2):138–41. Available from: https://linkinghub.elsevier.com/retrieve/pii/S1470211824053624 source
  • May combine two claims.

    Findingown data
    § Poster Type and Scores · p. 4

    The mean first impression score was 32.5 for #BetterPoster and 25.7 for classic format.

    “The mean scores for 'first impression' for #BetterPoster and classic format were 32.5 and 25.7, respectively.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingown data
    § First Impression by Poster Consensus Type · p. 4

    Higher scores for organization, poster design, or wordiness were associated with higher first impression scores.

    “Overall, the plot shows that the higher the score for organization, design, or wordiness, the higher the score for first impression”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Strengths and Limitations · p. 5

    The traffic for each poster during the whole conference was not measured systematically.

    “The traffic for each poster during the whole conference was not measured systematically.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data
    § Strengths and Limitations · p. 5

    The study did not conduct repeated evaluations of posters after a period of time, which could have provided insights into consistency.

    “Furthermore, we did not conduct repeated evaluations of the posters after a period of time, which could have provided insights into the consistency of our assessments.”
    Quote in the paperNot a substantial claim
  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation
    § Highlights · p. 2

    The #BetterPoster approach offers an alternative to classic formats, addressing the need for posters to be more visually engaging and reader friendly.

    “Specifically, the #BetterPoster approach offers an alternative to more classic formats, addressing the need for posters to be more visually engaging and reader friendly [3, 4].”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [3] Morrison M (2019) How to create a better research poster in less time (#betterposter Generation 1) [Internet]. YouTube. Available from: https://youtu.be/1RwJbhkCA58?si=l5wxZszTB8Ep7x_N. Accessed Apr 2024

      Not checked in the free run.

    • cites [4] Morrison M (2019) Better scientific poster [Internet]. Available from: osf.io/ef53g. Accessed Apr 2024

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data
    § Conclusion · p. 5

    The better poster format generally yielded higher scores for first impressions and overall visual appeal, but both formats can be effective if properly executed.

    “In conclusion, while the better poster format generally yielded higher scores for first impressions and overall visual appeal, both poster formats can be effective if properly executed.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Differences in poster scores between formats are attributable to the format rather than to content or other poster characteristics.
    • Rests on an assumption: The findings from the Danish Cancer Research Conference 2024 generalize to other scientific conferences.
  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductown data
    § Categorization of Poster Format · p. 2

    Poster format categorization was done independently by two raters, with disagreements resolved by joint reassessment to consensus.

    “The categorization was done independently by two raters (author LB and DEØ), and if there were disagreements, the poster was reassessed together, and consensus was reached.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion · p. 5

    Previous research advocates for structured formats such as IMRAD, which improves clarity and logical flow.

    “Previous research advocates for the use of structured formats, such as the IMRAD format (introduction, methods, results, and discussion), which improves clarity and aids in the logical flow of information and ensures proper conduct of research [6].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [6] Rose TM (2017) An illustrated guide to poster format [Internet]. Am J Pharm Educ 81(7):6423. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0002945923024397. Accessed Oct 2024 source
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Poster Type and Scores · p. 4

    Consensus for all posters was reached after a second joint assessment.

    “Consensus for all posters was reached with a second joint assessment (by author LB and DEØ).”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationreasoning
    § Abstract · p. 1

    Classic posters, although often more text-heavy, could also be effective when structured carefully.

    “Classic posters, although often more text-heavy, could also be effective when structured carefully.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Study design and conductown data
    § Interrater Reliability · p. 3

    Agreement for first impression, poster design, and wordiness was statistically significant.

    “Items with strong and moderate agreement were significant in agreement whereas the test for organization was not significant (table S2 for values).”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given
    § Strengths and Limitations · p. 5

    LB and DEØ had previously encouraged poster presenters to adopt the better poster format, which may have contributed to some degree of bias, although this influence did not extend to other raters.

    “Finally, LB and DEØ have previously encouraged poster presenters to adopt the better poster format, which may have contributed to some degree of bias, although this influence did not extend to the other raters.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductreasoning
    § Strengths and Limitations · p. 5

    The study's evaluations were subjective, and certain moderators' preference for the better poster format may have influenced scores, remaining a potential source of bias.

    “Firstly, the subjective nature of the evaluations, especially with certain moderators (including LB and DEØ) having a preference of the better poster format, may have influenced the scores and remains as a potential source of bias.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Interpretationreasoning
    § Discussion · p. 4

    Too many boxes eliminate guidance, text-heavy boxes make readers skip or skim, and boxes with no text leave audience puzzled.

    “Secondly, many posters used boxes to frame sections, and while this is a step towards guiding the audience, too many boxes will eliminate the guidance, just as a text heavy box will make readers skip or skim this part, and boxes with no text will leave the audience puzzled.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Interpretationown data
    § Discussion · p. 5

    Both formats have merits, and a well-executed poster borrows elements from both.

    “Therefore, based on the results from this study, we argue that both formats have their merits, and a well-executed poster borrows elements from both.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Poster Type and Scores · p. 4

    Initial consensus for categorizing poster format was 57%.

    “The initial consensus for categorizing the posters format was 57%.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Items Assessed · p. 2

    All raters received written information about the items and scales before the conference.

    “All raters had received written information about the items and scales before the conference.”
    Quote in the paperNot a substantial claim
  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given
    § Items Assessed · p. 2

    The four items were assessed by two to four raters per poster, with authors LB and DEØ rating all posters independently.

    “The four items were assessed by two to four raters per poster. Authors LB and DEØ rated all posters independently, and the moderators who moderated the poster presentations at the interactive poster session were invited to rate the posters in their track independently.”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim.

    Interpretationreasoning
    § Discussion · p. 4

    #BetterPoster posters could become disorganized, especially with random elements like excessive highlighting or ill-fitting images.

    “Posters with better poster format could become disorganized, especially if random elements like excessive highlighting or ill-fitting images were used.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion · p. 5

    A notable issue in poster preparation is the lack of formal training in communication and visualization of research.

    “In addition, a notable issue in poster preparation is the lack of formal training in communication and visualization of research with respect to different types of communication styles (scientific paper, news article, web pages, posters, etc.) [17-19].”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [17] Schade A (2018) Inverted pyramid: writing for comprehension [Internet]. Nielsen Norman Groupl. [cited 2024 Oct 23]. Available from: https://www.nngroup.com/articles/inverted-pyramid/ . Accessed Oct 2024

      Not checked in the free run.

    • cites [18] Minto B (2009) The pyramid principle, - logic in writing and thinking. 3rd revise. Essex, UK: Pearson Education Limited p. 177

      Not checked in the free run.

    • cites [19] Leventhal P, Reeves A. Manuscript writing. Med Writ [Internet]. 2012 Dec 18;21(4):314–6. Available from: http://www.tandfonline.com/doi/full/10.1179/2047480612Z.00000000076 source
28 of 28
  • The paper relies on this without support.

    First impression scores are a valid proxy for audience engagement or experience.

    “While the BP format improved audience engagement, excessive simplification or disorganization reduced its effectiveness.”
  • The paper relies on this without support.

    Differences in poster scores between formats are attributable to the format rather than to content or other poster characteristics.

    “The mean scores for 'first impression' for #BetterPoster and classic format were 32.5 and 25.7, respectively.”
  • The paper relies on this without support.

    The findings from the Danish Cancer Research Conference 2024 generalize to other scientific conferences.

    “While additional forums of varying sizes would increase the generalizability of our findings, our study offers a pragmatic and feasible approach to assess a recent paradigm shift in poster formats, contributing to evidence-based guidance for scientific dissemination.”
3 of 3

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. Spotlight on Events (2012) Nature [Internet]. Available from: https://www.nature.com/articles/nj0379. Accessed Oct 2024
  2. Minto B (2009) The pyramid principle, - logic in writing and thinking. 3rd revise. Essex, UK: Pearson Education Limited p. 177

Cited by

Sorted by citations

Authors
Bentsen, L., & Østergaard, D. E.
Affiliations
University of Copenhagen (DK)
Corresponding author
Østergaard, Daniella Elisabet
Venue
Journal of Cancer Education
Published
2025-04-02
Volume
41
Issue
1
Pages
49-54
ISSN
  • 0885-8195 (ISSN-L)
  • 1543-0154
OpenAlex
W4409147526
Cited by
5
References
12
Citations by year
2026: 3 · 2025: 2
FWCI
14.03
Type
journal article
Language
en
Open access
Locations
  • Journal of Cancer Education (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
Funders
Copenhagen University
Topics
  • Social Sciences › Arts and Humanities › History and Philosophy of Science › Academic Writing and Publishing
  • Social Sciences › Decision Sciences › Information Systems and Management › Academic Publishing and Open Access
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Data Visualization and Analytics
Keywords
research posters, poster design, visual communication, scientific dissemination, cancer research conferences, audience engagement, first impressions
MeSH terms
Communication, Congresses as Topic, Humans, Neoplasms, Information Dissemination, Posters as Topic

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-11-05.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-11-05.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review