OPOle Pütz

Ole Pütz

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Ole Pütz has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Top Cited Papers
    Open Access

    LLLMs: A Data-Driven Survey of Evolving Research on Limitations of Large Language Models

    by Aida Kostikova, Zhipin Wang, Deidamea Bajri, Ole Pütz, Benjamin Paaßen, Steffen Eger

    Large language model (LLM) research has grown rapidly, along with increasing concern about their limitations. In this survey, we conduct a data-driven, semi-automated review of research on limitations of LLMs ( LLLMs ) from 2022 to early 2025 using a bottom-up approach. From a corpus of 250,000 ACL and arXiv papers, we identify 14,648 relevant papers using keyword filtering, LLM-based classification, validated against expert labels, and topic clustering (via two approaches, HDBSCAN+BERTopic and LlooM). We find that the share of LLM-related papers increases over fivefold in ACL and nearly eightfold in arXiv between 2022 and 2025. Since 2022, LLLMs research grows even faster, reaching over 30% of LLM papers by 2025. Reasoning remains the most studied limitation, followed by generalization , hallucination , bias , and security . The distribution of topics in the ACL dataset stays relatively stable over time, while arXiv shifts toward security risks , alignment , hallucinations , knowledge editing , and multimodality . We offer a quantitative view of trends in LLLMs research and release a dataset of annotated abstracts and a validated methodology, available at: github.com/a-kostikova/LLLMs-Survey.


  • Open Access

    Co‐textual dopes: How LLMs produce contextually appropriate text in chat interactions with humans without access to context

    by Ole Pütz

    This paper asks how LLM‐based systems can produce text that is taken as contextually appropriate by humans without having seen text in its broader context. To understand how this is possible, context and co‐text have to be distinguished. Co‐text is input to LLMs during training and at inference as well as the primary resource of sense‐making for humans in interaction, collaboratively produced by both human and machine during chat. Systems can also passively participate in contextualization, insofar as cues are found in the co‐text and the user guides the system as to what is the appropriate context for them.


  • The quick termination of verbal conflicts expressed through disagreement

    by Ole Pütz, Hafsa Hassan

    This paper considers verbal conflicts at the workplace and asks how team members negotiate conflict termination. Our corpus consists of meetings where computer scientists discuss the progress of projects they are working on, a context where conflict resolution is crucial for the continuation of work. Following the literature, we define conflicts as continued disagreement. The analysis focuses on conflict termination formats that show whether a conflict can be resolved as well as episode length and disagreement mitigation that indicate the severity of the conflict. Power relations among participants are also discussed. Our results show that most conflicts end after two or three disagreements. Among the termination formats, submission is the most frequent, followed by stand-off and compromise. We further show that participants minimize conflicts by concluding them as soon as someone indicates their unwillingness to concede, whereas the pattern of conflict minimization varies with the termination format that participants negotiate.


  • Open Access

    Reciprocal Communication and Political Deliberation on Twitter

    by Robert Ackland, Felix Gumbert, Ole Pütz, Bryan Gertzel, Matthias Orlikowski

    Social media platforms such as Twitter/X are increasingly important for political communication but the empirical question as to whether such communication enhances democratic consensus building (the ideal of deliberative democracy) or instead contributes to societal polarisation via fostering of hate speech and “information disorders” such as echo chambers is worth exploring. Political deliberation involves reciprocal communication between users, but much of the recent research into politics on social media has focused on one-to-many communication, in particular the sharing and diffusion of information on Twitter via retweets. This paper presents a new approach to studying reciprocal political communication on Twitter, with a focus on extending network-analytic indicators of deliberation. We use the Twitter v2 API to collect a new dataset (#debatenight2020) of reciprocal communication on Twitter during the first debate of the 2020 US presidential election and show that a hashtag-based collection alone would have collected only 1% of the debate-related communication. Previous work into using social network analysis to measure deliberation has involved using discussion tree networks to quantify the extent of argumentation (maximum depth) and representation (maximum width); we extend these measures by explicitly incorporating reciprocal communication (via triad census) and the political partisanship of users (inferred via usage of partisan hashtags). Using these methods, we find evidence for reciprocal communication among partisan actors, but also point to a need for further research to understand what forms this communication takes.


  • Managing exactness and vagueness in computer science work: Programming and self-repair in meetings

    by Ole Pütz

    The formulation of computer algorithms requires the elimination of vagueness. This elimination of vagueness requires exactness in programming, and this exactness can be traced to meeting talk, where it intersects with the indexicality of expressions. This article is concerned with sequences in which a team of computer scientists discuss the functionality of prototypes that are already implemented or possibly to be implemented. The analysis focuses on self-repair because this is a practice where participants can be seen to orient to meanings of different expressions as alternatives. By using self-repair, the computer scientists show a concern with exact descriptions when they talk about existing functionality of their prototypes but not when they talk about potential future functionality. Instead, when participants talk about potential future functionality and attend to meanings during self-repair, they use vague expressions to indicate possibilities. Furthermore, when the computer scientists talk to external stakeholders, they indicate through hedges whenever their descriptions approximate already implemented technical functionality but do not describe it exactly. The article considers whether the code of working prototypes can be said to fix meanings of expressions and how we may account for human agency and non-human resistances during development.


  • Talking Collective Action

    by Ole Pütz

    An ethnographic study of anti-nuclear movement groups that both challenges assumptions of traditional social movement studies of strategic action and shows what can be gained through microanalysis of talk in meetings, this book advances social movement studies methodologically and theoretically through the application of a new method of sequential analysis. Drawing on both conversation analysis and objective hermeneutics, it builds on microanalysis to scale up from sequences of talk to meetings, from meetings to groups, and from groups to the anti-nuclear movement, thus addressing a common criticism of analyses of face-to-face interactions: that they fail to demonstrate how their findings are relevant for questions beyond the interaction itself and thus for a broader sociological audience. A demonstration of the ways in which strategic deliberations by activists are subject to dynamics of face-to-face interaction, Talking Collective Action shows how groups adopt different styles of planning to engage with their environment and affect the groups’ development over time. As such, it will appeal to social scientists with interests in social movements, organizations and conversation analysis.


  • How Strangers Initiate Conversations: Interactions on Public Trains in Germany

    by Ole Pütz

    This article considers how strangers who use public transportation initiate conversations and how disruptions of the transportation system affect interactions among strangers. How conversations are initiated has rarely been discussed in the literature because the majority of research takes the initiation of talk for granted. Building on Goffman, the article tests two hypotheses that explain how strangers initiate conversations. The first hypothesis states that travelers rely on interactional rituals if they have to talk with others because of a rule against opening talk with strangers, a rule that can be relaxed if travelers are faced with disruptive events. The second hypothesis states that a conversation can be initiated without introductory remarks if a traveler’s focus of attention is discernible to another traveler, irrespective of the circumstances travelers find themselves in. I argue that the latter hypothesis better explains how strangers initiate conversations and discuss how this finding may be generalized.


  • From Non-Places to Non-Events

    by Ole Pütz

    In this article I raise the question of how individuals avoid interaction in spite of bodily proximity. I consider the empirical case of the airport security checkpoint. Drawing on participant observation of nine different airports, my research focuses on the security checkpoints of a small German airport. I suggest that the security checkpoint functions as a junction of security and mobility where screeners and travelers cannot avoid physical contact. Guided by the insights of Goffman and Augé, my analysis reveals that practices akin to civil inattention, which are based on the avoidance of interaction, create the experience of a non-place through “non-events.” A precondition for non-events is the standardization of procedures at the checkpoint, which creates a visual and sequential order. Strict procedural rules allow screeners and travelers to disconnect the body of the traveler temporarily from the traveler as an individual. Screeners and travelers practically achieve this disconnection by minimizing face-to-face interaction, thereby avoiding problematic definitions of the situation. This practice, however, results in discrimination against travelers who are unable or unwilling to present their body as a normalized and passive body. My findings thus point to the ability of organizationally framed settings, like the security checkpoint, to deviate from public life, while also addressing the limits of such a deviation. In concluding, I discuss the utility of the concept of non-places along with the merits of interactional analysis for understanding consequences of technological and procedural change. I also consider the efficiency of security practices from a dramaturgical perspective.