Pith. sign in

REVIEW 3 major objections 4 minor 119 references

A Computational Framework to Identify Self-Aspects in Text

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper proposes that the Self's multiple aspects—bodily, narrative, agentive, social, and minimal—leave identifiable traces in language, and that a computational framework built from an ontology, annotated datasets, and a mix of NLP…

desk verdict A well-structured Ph.D. proposal whose only pilot is undercut by LLM labels inside the cross-validation set; the plan is worth engaging, but the empirical claim is not yet supported. read the letter →

arxiv 2507.13115 v1 pith:6R6DNSS3 submitted 2025-07-17 cs.CL

Jaya Caporusso , Matthew Purver , Senja Pollak This is my paper
classification cs.CL
keywords Self-aspectsontologynaturallanguageprocessingtextclassificationphenomenologymentalhealthlinguisticmarkerslargemodels
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proposes that the Self should be treated as a structured set of partially independent aspects—bodily, narrative, agentive, social, and minimal—and that these aspects leave identifiable traces in ordinary language. Its plan is to turn that idea into a working framework: build a detailed ontology of aspects, elements, and modes, annotate a diverse set of texts against it, and then train and compare conventional classifiers, large language models, and embedding-based retrieval systems. The only completed experiment is a pilot that labels one aspect, the Social Self, in 1,473 diary sub-entries, where a support-vector machine using psychology-grounded word-category features reaches macro F1 of 0.83. If the full framework delivers on this, it would give mental-health researchers and empirical phenomenologists an automated way to locate passages where agency, embodiment, narrative identity, or social selfhood is at stake.

What carries the argument

The load-bearing object is the Self-aspect ontology: a structured inventory of the Self broken into aspects (Bodily Self, Narrative Self, Self as Agent, Minimal Self, Social Self), elements within each aspect (for example, body ownership and body awareness), and modes of each element (for instance, weak or strong ownership), each supplied with definitions and positive and negative textual examples. This ontology does the work of converting a philosophical construct into a concrete annotation scheme, because it defines exactly what label a text chunk can receive. The pilot's other piece of machinery is a set of psychologically grounded word-category features derived from prior work on which linguistic markers correlate with each Self-aspect; those features give a simple classifier an interpretable basis for distinguishing, say, a social-self utterance from one that is not. The paper also plans a common evaluation protocol across models, using an established statistical comparison method over at least five datasets.

What would settle it

An empirical test would be to run the planned annotation on the full ontology with three or more annotators over the proposed heterogeneous corpora: if any core Self-aspect beyond the pilot's Social Self fails to reach inter-annotator agreement clearly above chance after guideline refinement, then that aspect is not reliably detectable in text and the framework's central premise is falsified.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is that a theoretically grounded ontology of Self-aspects can function as a text-classification scheme: each aspect, element, and mode can be reliably labelled in text, and models trained on those labels can generalise beyond the training material. The pilot is offered as the concrete evidence for this feasibility: an SVM that reads diary sub-entries through psychological word-category features classifies the presence of the Social Self with macro precision, recall, and F1 each at 0.83, with inter-annotator agreement of 0.80 between human annotators and 0.84–0.89 between human and machine annotators. The paper acknowledges that this supports only binary classification of a single Self-aspect in English diary text from one source, so the broader claim about all five core aspects remains a proposal with a proof of concept, not an established result.

Load-bearing premise

The framework assumes that the ontology of Self-aspects is well defined enough that multiple annotators can reliably label arbitrary text chunks with the same aspect, element, and mode; if the labels are noisy or inconsistent, every downstream model inherits that noise.

Editorial extensions

If this is right

  • If the framework works across the planned dataset types, clinicians could screen patient narratives and diaries for changes in sense of agency, body ownership, or narrative self-continuity.
  • Phenomenologists could use the models to flag passages in interview transcripts where a specific Self-aspect or mode is expressed, supporting micro-phenomenological analysis.
  • The pilot's hybrid human-LLM annotation pipeline, with synthetic labels kept out of evaluation, would allow larger training sets without contaminating measured performance.
  • The planned bias evaluation, using minimal sentence pairs differing only in socially salient variables, would check whether model predictions are invariant to gender or race markers.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: because the paper itself says the Self-aspects are not mutually exclusive, a natural test is to allow annotators to assign multiple labels to one chunk and see how often the ontology's categories co-occur; the framework may need multi-label or span-based prediction rather than single-label classification.
  • Editorial inference: the strongest generalisability test would be a small multilingual annotation study, since pronoun use and agency phrasing vary across languages and cultures and the pilot is English-only.
  • Editorial inference: the practical value of the framework will ultimately be judged by whether domain experts, not just classifiers, find the flagged passages useful; a user study with phenomenologists or clinicians is a natural next step beyond accuracy metrics.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. This manuscript presents a PhD proposal to build a computational framework for identifying Self-aspects in text, combining an ontology, annotated datasets, and several model families. It describes research objectives, a work-in-progress ontology, a dataset and annotation plan, four desiderata, proposed modelling approaches, and an evaluation strategy. An appendix reports a pilot study on binary Social Self classification on 1,473 diary sub-entries, with an SVM on LIWC features achieving macro-F1=0.83.

Significance. If implemented as planned, the framework would address a real gap in NLP by moving beyond coarse LIWC-style markers toward theory-grounded detection of phenomenological Self-aspects. The paper is clearly organized, explicitly acknowledges its preliminary status, and proposes a sensible desiderata set with a bias-evaluation plan and a multi-dataset statistical comparison. The pilot, although currently limited, is a useful exploratory step, and Section 5.1's commitment to excluding LLM-based annotations from evaluation is a good methodological intention. The full contribution is not yet realized; at this stage the paper's value is as a roadmap.

major comments (3)
  1. [Appendix A.1–A.3 and Section 5.1] The pilot's 10-fold cross-validation uses a gold standard that mixes manual labels with labels produced by Gemma2, and the text does not state whether LLM-labelled instances were excluded from the test folds. This conflicts with the clean separation promised in Section 5.1, and it means the reported macro-F1 of 0.83 for the best SVM can be read as agreement with the Gemma2 annotation style rather than with a human gold standard. Please re-run the pilot with a human-only held-out set, or at least report the proportion of LLM-labelled instances and verify their absence from test folds, and temper the A.4 statement that the pilot 'demonstrates' reliability.
  2. [Section 4] The ontology is work-in-progress, and the paper acknowledges that Self-aspects are not mutually exclusive and that aspects such as sense of agency can apply to other aspects. The planned labelling protocol does not specify whether annotations are single-label or multi-label, nor how annotators should handle co-occurring aspects; without this, the gold-standard labels (RO2) and every downstream model inherit an ambiguous target. Please provide an explicit annotation rule for overlapping aspects and a worked example from the future guidelines.
  3. [Sections 6 and 8.1] The proposed evaluation compares models on four desiderata, but only accuracy (D3) is given a standard metric; interpretability (D1), ground-truth adherence (D2), and computational cost (D4) are not operationalized. To make the planned Demšar (2006) comparison meaningful, specify measurable proxies, such as the proportion of decisions traceable to a retrieved example, runtime or energy per instance, and agreement with a pre-registered annotation protocol.
minor comments (4)
  1. [Section 2.1] The phrase 'focuses onI-talk' is missing a space and should read 'focuses on I-talk'.
  2. [Section 4] The example excerpts label categories such as 'Sense of Ownership and Sense of Control' and 'Thoughts' that are not introduced in the preceding aspect list; please clarify how these map to the defined aspects and their elements.
  3. [Section 8.2] The word 'desiderable' appears where 'desirable' is intended.
  4. [Appendix A.3] The pairwise Wilcoxon signed-rank tests are reported only as 'adjusted p = 0.03 for several comparisons'; please list which model pairs reached significance.

Circularity Check

1 steps flagged · score 6.0 of 10

Pilot F1 is evaluated against a mixed human/LLM gold set, so the only empirical support for Self-aspect detectability partly reduces to agreement with the LLM annotation pipeline.

  1. other [Appendix A.1-A.3; cf. Section 5.1]
    "We employed a publicly available dataset of 1,473 diary sub-entries (Li and Parikh, 2019), which we augmented with binary annotations for SS. Annotation combined manual labelling and automated classification using three versions of Gemma2... We trained and evaluated six models using 10-fold cross-validation... The best-performing model was the SVM trained on LIWC features, achieving a macro-averaged precision of 0.83 (STD = 0.03), recall of 0.83 (STD = 0.03), and F1-score of 0.83 (STD = 0.03) across 10 folds. Section 5.1: 'LLM-based annotations will be used to augment training data ..."

    The pilot's 10-fold cross-validation trains and tests on the same augmented set, and the paper does not report holding out the Gemma2-labelled instances from the test folds. Since Section 5.1 explicitly classes LLM annotations as synthetic data that must be excluded from evaluation 'to avoid circularity', any F1 computed over the mixed set measures, for those instances, the SVM's agreement with the Gemma2 annotation style rather than with an independent human gold standard. The reported F1=0.83 therefore cannot by itself support the pilot's claim that interpretable models 'reliably identify' the Social Self in text; it is, by construction, partly an agreement score with the annotation model.

full rationale

The paper is largely a Ph.D. proposal, and most of its derivation chain (ontology construction, annotation plans, model desiderata, planned Demšar-based evaluation) has independent content. Self-citations to Caporusso et al. (2024) document prior ontology and LIWC-correlation work rather than serving as an unverified uniqueness premise, so they are not by themselves circular. The circularity is concentrated in Appendix A, which is the only empirical support for the feasibility claim. The pilot's gold standard is a mixture of manual labels and Gemma2-generated labels, and the 10-fold cross-validation is run over this mixed set without any stated holdout of synthetic instances, so the reported F1 is partly an agreement score with the LLM annotation pipeline rather than a human-ground-truth measure. This is precisely the circularity the paper says it will avoid in Section 5.1. The planned future evaluation, if it excludes synthetic labels from test folds, would correct this; the current paper does not claim that exclusion for the pilot. The rest of the proposed framework is not circular, but the pilot is load-bearing for the paper's initial feasibility claim.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The proposal rests on domain assumptions about the textual detectability of Self-aspects and the reliability of a consensus-based ontology. No formal axioms or invented entities are introduced. The pilot uses standard statistical tests and standard classifiers.

assumptions (4)
  • domain assumption Self-aspects are reflected in the language a person uses.
    The entire framework presupposes that phenomenological and psychological Self-aspects leave detectable textual traces; invoked in the Abstract and Introduction.
  • domain assumption The proposed ontology's Self-aspect categories, elements, and modes are well-defined and consistently annotatable from text chunks.
    Section 4 acknowledges non-exclusivity and disciplinary disagreements; if annotation labels are unreliable, all downstream models inherit the noise.
  • domain assumption Existing NLP methods (SVMs, LLMs, embeddings) can be adapted to this fine-grained classification task.
    Sections 7.1 to 7.4 assume the modeling approaches will transfer; the pilot provides weak support for one aspect.
  • standard math Standard statistical tests (Friedman, Wilcoxon) are valid for comparing classifiers across datasets.
    Used in Appendix A.3 to support the significance of the pilot results.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Computational Framework to Identify Self-Aspects in Text." pith.science (2026). https://pith.science/paper/6R6DNSS3

@misc{pith2026250713115,
  author       = {Pith},
  title        = {Pith review of: A Computational Framework to Identify Self-Aspects in Text},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6R6DNSS3}},
  note         = {Machine review of arXiv:2507.13115}
}
read the original abstract

This Ph.D. proposal introduces a plan to develop a computational framework to identify Self-aspects in text. The Self is a multifaceted construct and it is reflected in language. While it is described across disciplines like cognitive science and phenomenology, it remains underexplored in natural language processing (NLP). Many of the aspects of the Self align with psychological and other well-researched phenomena (e.g., those related to mental health), highlighting the need for systematic NLP-based analysis. In line with this, we plan to introduce an ontology of Self-aspects and a gold-standard annotated dataset. Using this foundation, we will develop and evaluate conventional discriminative models, generative large language models, and embedding-based retrieval approaches against four main criteria: interpretability, ground-truth adherence, accuracy, and computational efficiency. Top-performing models will be applied in case studies in mental health and empirical phenomenology.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

119 extracted references · 59 canonical work pages

  1. [1]

    Jonathan M Adler. 2012. Living into the story: agency and coherence in a longitudinal study of narrative identity development and mental health over the course of psychotherapy. Journal of personality and social psychology, 102(2):367

    2012

  2. [2]

    Jonathan M Adler, Lauren M Skalina, and Dan P McAdams. 2008. The narrative reconstruction of psychotherapy and psychological health. Psychotherapy research, 18(6):719--734

    2008

  3. [3]

    Muhammad Aurangzeb Ahmad, Carly Eckert, and Ankur Teredesai. 2018. Interpretable machine learning in healthcare. In Proceedings of the 2018 ACM international conference on bioinformatics, computational biology, and health informatics, pages 559--560

    2018

  4. [4]

    David Alvarez-Melis and Tommi S Jaakkola. 2018. On the robustness of interpretability methods. arXiv preprint arXiv:1806.08049

    arXiv 2018

  5. [5]

    Julia Amann, Alessandro Blasimme, Effy Vayena, Dietmar Frey, Vince I Madai, and Precise4Q Consortium. 2020. Explainability for artificial intelligence in healthcare: a multidisciplinary perspective. BMC medical informatics and decision making, 20:1--9

    2020

  6. [6]

    Patrik Aspers. 2009. Empirical phenomenology: A qualitative research approach (the cologne seminars). Indo-pacific journal of phenomenology, 9(2)

    2009

  7. [7]

    Michael Bamberg. 2008. Considering counter narratives. In Considering counter-narratives: Narrating, resisting, making sense, pages 351--371. John Benjamins Publishing Company

    2008

  8. [8]

    Roy F Baumeister, Arlene M Stillwell, and Todd F Heatherton. 1994. Guilt: an interpersonal approach. Psychological bulletin, 115(2):243

    1994

Show all 119 references
  1. [9]

    Yoshua Bengio, R \'e jean Ducharme, Pascal Vincent, and Christian Jauvin. 2003. A neural probabilistic language model. Journal of machine learning research, 3(Feb):1137--1155

    2003

  2. [10]

    Jos \'e Luis Berm \'u dez. 2018. The bodily self: Selected essays. MIT Press

    2018

  3. [11]

    Christopher M Bishop and Nasser M Nasrabadi. 2006. Pattern recognition and machine learning, volume 4. Springer

    2006

  4. [12]

    Lasse Bohlen, Julian Rosenberger, Patrick Zschech, and Mathias Kraus. 2024. Leveraging interpretable machine learning in intensive care. Annals of Operations Research, pages 1--40

    2024

  5. [13]

    Tolga Bolukbasi, Kai-Wei Chang, James Y Zou, Venkatesh Saligrama, and Adam T Kalai. 2016. Man is to computer programmer as woman is to homemaker? debiasing word embeddings. Advances in neural information processing systems, 29

    2016

  6. [14]

    Ryan L Boyd, Ashwini Ashokkumar, Sarah Seraj, and James W Pennebaker. 2022. The development and psychometric properties of liwc-22. Austin, TX: University of Texas at Austin, 10:1--47

    2022

  7. [15]

    Ryan L Boyd and H Andrew Schwartz. 2021. Natural language analysis and the psychology of verbal behavior: The past, present, and future states of the field. Journal of Language and Social Psychology, 40(1):21--41

    2021

  8. [16]

    Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. 2020. Language models are few-shot learners. Advances in neural information processing systems, 33:1877--1901

    2020

  9. [17]

    Jerome Seymour Bruner. 2003. Making stories: Law, literature, life. Harvard University Press

    2003

  10. [18]

    Jaya Caporusso. 2022. Dissolution experiences and the experience of the self: an empirical phenomenological investigation (master’s thesis). university of vienna. Advisor: Assist. Prof. Dr. Maja Smrdu

    2022

  11. [19]

    Jaya Caporusso, Boshko Koloski, Ma s a Rebernik, Senja Pollak, and Matthew Purver. 2024. A phenomenologically-inspired computational analysis of self-categories in text. In Proceedings of JADT 2024

    2024

  12. [20]

    Jaya Caporusso, Thi Hong Hanh Tran, and Senja Pollak. 2023. https://aclanthology.org/2023.ltedi-1.26/ IJS @ LT - EDI : Ensemble approaches to detect signs of depression from social media text . In Proceedings of the Third Workshop on Language Technology for Equality, Diversity...

    2023

  13. [21]

    Daniel Cer, Yinfei Yang, Sheng-yi Kong, Nan Hua, Nicole Limtiaco, Rhomni St John, Noah Constant, Mario Guajardo-Cespedes, Steve Yuan, Chris Tar, et al. 2018. Universal sentence encoder. arXiv preprint arXiv:1803.11175

    2018 arXiv

  14. [22]

    Muhammad Waqas Anjum Ch and Waqas Arshad Cheema. 2018. A study of content based methods for author profiling in multiple genres. International Journal of Scientific Engineering Research, 9(9):322--327

    2018

  15. [23]

    Ahmad Chaddad, Jihao Peng, Jian Xu, and Ahmed Bouridane. 2023. Survey of explainable ai techniques in healthcare. Sensors, 23(2):634

    2023

  16. [24]

    Chakravarthi, B

    Bharathi R. Chakravarthi, B. Bharathi, Joephine Griffith, Kalika Bali, and Paul Buitelaar, editors. 2023. https://aclanthology.org/2023.ltedi-1.0/ Proceedings of the Third Workshop on Language Technology for Equality, Diversity and Inclusion . INCOMA Ltd., Shoumen, Bulgaria, V...

    2023

  17. [25]

    Mara Chinea-Rios, Thomas M \"u ller, Gretel Liz De la Pe \ n a Sarrac \'e n, Francisco Rangel, and Marc Franco-Salvador. 2022. Zero and few-shot learning for author profiling. In International Conference on Applications of Natural Language to Information Systems, pages 333--34...

    2022

  18. [26]

    Nello Cristianini and John Shawe-Taylor. 2000. An introduction to support vector machines and other kernel-based learning methods. Cambridge university press

    2000

  19. [27]

    Janez Dem s ar. 2006. Statistical comparisons of classifiers over multiple data sets. Journal of Machine learning research, 7(Jan):1--30

    2006

  20. [28]

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human...

    2019

  21. [29]

    Finale Doshi-Velez and Been Kim. 2017. Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608

    2017 arXiv

  22. [30]

    Xiaowei Du and Yunmei Sun. 2022. Linguistic features and psychological states: A machine-learning based approach. Frontiers in psychology, 13:955850

    2022

  23. [31]

    Johannes C Eichstaedt, Robert J Smith, Raina M Merchant, Lyle H Ungar, Patrick Crutchley, Daniel Preo t iuc-Pietro, David A Asch, and H Andrew Schwartz. 2018. Facebook language predicts depression in medical records. Proceedings of the National Academy of Sciences, 115(44):112...

    2018

  24. [32]

    Christopher Ifeanyi Eke, Azah Anir Norman, Liyana Shuib, and Henry Friday Nweke. 2019. A survey of user profiling: State-of-the-art, challenges, and solutions. IEEE Access, 7:144907--144924

    2019

  25. [33]

    Mohamad El Haj, Pascal Antoine, Jean Louis Nandrino, and Dimitrios Kapogiannis. 2015. Autobiographical memory decline in alzheimer’s disease, a theoretical and clinical overview. Ageing research reviews, 23:183--192

    2015

  26. [34]

    Mohammad Ennab and Hamid Mcheick. 2024. Enhancing interpretability and accuracy of ai models in healthcare: a comprehensive review on challenges and future directions. Frontiers in Robotics and AI, 11:1444763

    2024

  27. [35]

    Bojan Evkoski and Senja Pollak. 2023. Xai in computational linguistics: Understanding political leanings in the slovenian parliament. arXiv preprint arXiv:2305.04631

    2023 arXiv

  28. [36]

    Mehwish Fatima, Komal Hasan, Saba Anwar, and Rao Muhammad Adeel Nawab. 2017. Multilingual author profiling on facebook. Information Processing & Management, 53(4):886--904

    2017

  29. [37]

    Mark Freeman. 2009. Hindsight: The promise and peril of looking backward. Oxford University Press

    2009

  30. [38]

    Shaun Gallagher. 2000. Philosophical conceptions of the self: implications for cognitive science. Trends in cognitive sciences, 4(1):14--21

    2000

  31. [39]

    Xiaoquan Gao, Sabriya Alam, Pengyi Shi, Franklin Dexter, and Nan Kong. 2023. Interpretable machine learning models for hospital readmission prediction: a two-step extracted regression tree approach. BMC medical informatics and decision making, 23(1):104

    2023

  32. [40]

    Ian Goodfellow, Yoshua Bengio, Aaron Courville, and Yoshua Bengio. 2016. Deep learning, volume 1. MIT press Cambridge

    2016

  33. [41]

    Sharath Chandra Guntuku, Rachelle Schneider, Arthur Pelullo, Jami Young, Vivien Wong, Lyle Ungar, Daniel Polsky, Kevin G Volpp, and Raina Merchant. 2019. Studying expressions of loneliness in individuals using twitter: an observational study. BMJ open, 9(11):e030355

    2019

  34. [42]

    Tilmann Habermas and Christin K \"o ber. 2015. Autobiographical reasoning in life narratives buffers the effect of biographical disruptions on the sense of self-continuity. Memory, 23(5):664--674

    2015

  35. [43]

    Yaakov HaCohen-Kerner. 2022. Survey on profiling age and gender of text authors. Expert Systems with Applications, 199:117140

    2022

  36. [44]

    Henri \"e tte Doroth \'e e Heering, Saskia Goedhart, Richard Bruggeman, Wiepke Cahn, Lieuwe de Haan, Ren \'e S Kahn, Carin J Meijer, Inez Myin-Germeys, Jim van Os, and Durk Wiersma. 2016. Disturbed experience of self: psychometric analysis of the self-experience lifetime frequ...

    2016

  37. [45]

    Tine Holm, Dorthe Kirkegaard Thomsen, and Vibeke Bliksted. 2016. Life story chapters and narrative self-continuity in patients with schizophrenia. Consciousness and cognition, 45:60--74

    2016

  38. [46]

    Dirk Hovy and Shrimai Prabhumoye. 2021. Five sources of bias in natural language processing. Language and linguistics compass, 15(8):e12432

    2021

  39. [47]

    Jeremy Howard and Sebastian Ruder. 2018. Universal language model fine-tuning for text classification. arXiv preprint arXiv:1801.06146

    2018 arXiv

  40. [48]

    Baixiang Huang, Canyu Chen, and Kai Shu. 2024. Can large language models identify authorship? arXiv preprint arXiv:2403.08213

    2024 arXiv

  41. [49]

    Baixiang Huang, Canyu Chen, and Kai Shu. 2025. Authorship attribution in the era of llms: Problems, methodologies, and challenges. ACM SIGKDD Explorations Newsletter, 26(2):21--43

    2025

  42. [50]

    Clayton Hutto and Eric Gilbert. 2014. Vader: A parsimonious rule-based model for sentiment analysis of social media text. In Proceedings of the international AAAI conference on web and social media, volume 8, pages 216--225

    2014

  43. [51]

    Jeff Jaeger, Katie M Lindblom, Kelly Parker-Guilbert, and Lori A Zoellner. 2014. Trauma narratives: It’s what you say, not how you say it. Psychological Trauma: Theory, Research, Practice, and Policy, 6(5):473

    2014

  44. [52]

    Di Jin, Zhijing Jin, Joey Tianyi Zhou, and Peter Szolovits. 2020. Is bert really robust? a strong baseline for natural language attack on text classification and entailment. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 8018--8025

    2020

  45. [53]

    Vladimir Karpukhin, Barlas Oguz, Sewon Min, Patrick SH Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih. 2020. Dense passage retrieval for open-domain question answering. In EMNLP (1), pages 6769--6781

    2020

  46. [54]

    Svetlana Kiritchenko and Saif M Mohammad. 2018. Examining gender and race bias in two hundred sentiment analysis systems. arXiv preprint arXiv:1805.04508

    2018 arXiv

  47. [55]

    Tine Kolenik, G \"u nter Schiepek, and Matja z Gams. 2024. Computational psychotherapy system for mental health prediction and behavior change with a conversational agent. Neuropsychiatric Disease and Treatment, pages 2465--2498

    2024

  48. [56]

    Tia Križan, Luka Oprešnik, and Jaya Caporusso. 2025. Toward an ontology of the self: A theoretical framework. In Proceedings of the MEi:CogSci Conference, volume 19

    2025

  49. [57]

    Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. 2015. Deep learning. nature, 521(7553):436--444

    2015

  50. [58]

    u ttler, Mike Lewis, Wen-tau Yih, Tim Rockt \

    Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich K \"u ttler, Mike Lewis, Wen-tau Yih, Tim Rockt \"a schel, et al. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in neural information proc...

    2020

  51. [59]

    X Alice Li and Devi Parikh. 2019. Lemotif: An affective visual journal using deep neural networks. arXiv preprint arXiv:1903.07766

    2019 arXiv

  52. [60]

    Zachary C Lipton. 2018. The mythos of model interpretability: In machine learning, the concept of interpretability is both important and slippery. Queue, 16(3):31--57

    2018

  53. [61]

    Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692

    2019 arXiv

  54. [62]

    Sheng-Chieh Lu, Christine L Swisher, Caroline Chung, David Jaffray, and Chris Sidey-Gibbons. 2023. On the importance of interpretable machine learning predictions to inform clinical decision making in oncology. Frontiers in oncology, 13:1129380

    2023

  55. [63]

    Scott M Lundberg and Su-In Lee. 2017. A unified approach to interpreting model predictions. Advances in neural information processing systems, 30

    2017

  56. [64]

    Paul Henry Lysaker and John Timothy Lysaker. 2002. Narrative structure in psychosis: Schizophrenia and disruptions in the dialogical self. Theory & Psychology, 12(2):207--220

    2002

  57. [65]

    Dan P McAdams. 2001. The psychology of life stories. Review of general psychology, 5(2):100--122

    2001

  58. [66]

    Seifeddine Mechti, Nabil Khoufi, and Lamia Hadrich Belguith. 2020. Improving native language identification model with syntactic features: Case of arabic. In Intelligent Systems Design and Applications: 18th International Conference on Intelligent Systems Design and Applicatio...

    2020

  59. [67]

    Marishka M Mehta, Soojung Na, Xiaosi Gu, James W Murrough, and Laurel S Morris. 2023. Reward-related self-agency is disturbed in depression and anxiety. PloS one, 18(3):e0282727

    2023

  60. [68]

    o gele, Hartmut Sch \

    Matthias Michal, Bettina Reuchlein, Julia Adler, Iris Reiner, Manfred E Beutel, Claus V \"o gele, Hartmut Sch \"a chinger, and Andre Schulz. 2014. Striking discrepancy of anomalous body experiences with normal interoceptive accuracy in depersonalization-derealization disorder....

    2014

  61. [69]

    Brent Mittelstadt. 2019. Principles alone cannot guarantee ethical ai. Nature machine intelligence, 1(11):501--507

    2019

  62. [70]

    Hadi Mohammadi, Ayoub Bagheri, Anastasia Giachanou, and Daniel L Oberski. 2025. Explainability in practice: A survey of explainable nlp across various domains. arXiv preprint arXiv:2502.00837

    2025 arXiv

  63. [71]

    Christoph Molnar. 2020. Interpretable machine learning. Lulu. com

    2020

  64. [72]

    James W Moore. 2016. What is the sense of agency and why does it matter? Frontiers in psychology, 7:1272

    2016

  65. [73]

    Kevin P Murphy. 2012. Machine learning: a probabilistic perspective. MIT press

    2012

  66. [74]

    Jinkyung Na and Incheol Choi. 2009. Culture and first-person pronouns. Personality and Social Psychology Bulletin, 35(11):1492--1499

    2009

  67. [75]

    Ohad Nave, Fynn-Mathis Trautwein, Yochai Ataria, Yair Dor-Ziderman, Yoav Schweitzer, Stephen Fulder, and Aviva Berkovich-Ohana. 2021. Self-boundary dissolution in meditation: A phenomenological investigation. Brain Sciences, 11(6):819

    2021

  68. [76]

    Matthew M Nour, Lisa Evans, David Nutt, and Robin L Carhart-Harris. 2016. Ego-dissolution and psychedelics: validation of the ego-dissolution inventory (edi). Frontiers in human neuroscience, 10:190474

    2016

  69. [77]

    Sarra Ouni, Fethi Fkih, and Mohamed Nazih Omri. 2021. Toward a new approach to author profiling based on the extraction of statistical features. Social Network Analysis and Mining, 11(1):59

    2021

  70. [78]

    Sarra Ouni, Fethi Fkih, and Mohamed Nazih Omri. 2023 a . Novel semantic and statistic features-based author profiling approach. Journal of Ambient Intelligence and Humanized Computing, 14(9):12807--12823

    2023

  71. [79]

    Sarra Ouni, Fethi Fkih, and Mohamed Nazih Omri. 2023 b . A survey of machine learning-based author profiling from texts analysis in social networks. Multimedia Tools and Applications, 82(24):36653--36686

    2023

  72. [80]

    Josef Parnas, Paul M ller, Tilo Kircher, J rgen Thalbitzer, Lennart Jansson, Peter Handest, and Dan Zahavi. 2005. Ease: examination of anomalous self-experience. Psychopathology, 38(5):236

    2005

  73. [81]

    James W Pennebaker and Sandra K Beall. 1986. Confronting a traumatic event: toward an understanding of inhibition and disease. Journal of abnormal psychology, 95(3):274

    1986

  74. [82]

    James W Pennebaker and Martha E Francis. 1996. Cognitive, emotional, and language processes in disclosure. Cognition & emotion, 10(6):601--626

    1996

  75. [83]

    James W Pennebaker, Matthias R Mehl, and Kate G Niederhoffer. 2003. Psychological aspects of natural language use: Our words, our selves. Annual review of psychology, 54(1):547--577

    2003

  76. [84]

    Claire Petitmengin, Anne Remillieux, and Camila Valenzuela-Moguillansky. 2019. Discovering the structures of lived experience: Towards a micro-phenomenological analysis method. Phenomenology and the Cognitive Sciences, 18(4):691--730

    2019

  77. [85]

    Valeria I Petkova, Malin Bj \"o rnsdotter, Giovanni Gentile, Tomas Jonsson, Tie-Qiang Li, and H Henrik Ehrsson. 2011. From part-to whole-body ownership in the multisensory brain. Current Biology, 21(13):1118--1122

    2011

  78. [86]

    Alec Radford, Karthik Narasimhan, Tim Salimans, Ilya Sutskever, et al. 2018. Improving language understanding by generative pre-training

    2018

  79. [87]

    Matthew Ratcliffe. 2014. Experiences of depression: A study in phenomenology. OUP Oxford

    2014

  80. [88]

    Nils Reimers and Iryna Gurevych. 2019. Sentence-bert: Sentence embeddings using siamese bert-networks. arXiv preprint arXiv:1908.10084

    2019 arXiv

  81. [89]

    Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. Why should i trust you? explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, pages 1135--1144

    2016

  82. [90]

    Stephanie Rude, Eva-Maria Gortner, and James Pennebaker. 2004. Language use of depressed and depression-vulnerable college students. Cognition & Emotion, 18(8):1121--1133

    2004

  83. [91]

    Cynthia Rudin. 2019. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature machine intelligence, 1(5):206--215

    2019

  84. [92]

    Rupsa Saha, Ole-Christoffer Granmo, and Morten Goodwin. 2023. Using tsetlin machine to discover interpretable rules in natural language processing applications. Expert Systems, 40(4):e12873

    2023

  85. [93]

    Rupsa Saha, Ole-Christoffer Granmo, Vladimir I Zadorozhny, and Morten Goodwin. 2022. A relational tsetlin machine with applications to natural language understanding. Journal of Intelligent Information Systems, pages 1--28

    2022

  86. [94]

    H Andrew Schwartz, Johannes C Eichstaedt, Margaret L Kern, Lukasz Dziurzynski, Stephanie M Ramones, Megha Agrawal, Achal Shah, Michal Kosinski, David Stillwell, Martin EP Seligman, et al. 2013. Personality, gender, and age in the language of social media: The open-vocabulary a...

    2013

  87. [95]

    Andrea Serino, Adrian Alsmith, Marcello Costantini, Alisa Mandrigin, Ana Tajadura-Jimenez, and Christophe Lopez. 2013. Bodily ownership and self-location: components of bodily self-consciousness. Consciousness and cognition, 22(4):1239--1252

    2013

  88. [96]

    Elizabeth A Sharpless. 1985. Identity formation as reflected in the acquisition of person pronouns. Journal of the American Psychoanalytic Association, 33(4):861--885

    1985

  89. [97]

    Mark Siderits, Evan Thompson, and Dan Zahavi. 2013. Self, no self?: Perspectives from analytical, phenomenological, and Indian traditions. OUP Oxford

    2013

  90. [98]

    Mauricio Sierra and German E Berrios. 2000. The cambridge depersonalisation scale: a new instrument for the measurement of depersonalisation. Psychiatry research, 93(2):153--164

    2000

  91. [99]

    Almog Simchon, Britt Hadar, and Michael Gilead. 2023. A computational text analysis investigation of the relation between personal and linguistic agency. Communications Psychology, 1(1):23

    2023

  92. [100]

    Vinitra Swamy, Syrielle Montariol, Julian Blackwell, Jibril Frej, Martin Jaggi, and Tanja K \"a ser. 2025. Intrinsic user-centric interpretability through global mixture of experts. In The Thirteenth International Conference on Learning Representations

    2025

  93. [101]

    Shogo Tanaka. 2018. What is it like to be disconnected from the body?: A phenomenological account of disembodiment in depersonalization/derealization disorder. Journal of Consciousness Studies, 25(5-6):239--262

    2018

  94. [102]

    Yla R Tausczik and James W Pennebaker. 2010. The psychological meaning of words: Liwc and computerized text analysis methods. Journal of language and social psychology, 29(1):24--54

    2010

  95. [103]

    Gemma Team, Morgane Riviere, Shreya Pathak, Pier Giuseppe Sessa, Cassidy Hardin, Surya Bhupatiraju, L \'e onard Hussenot, Thomas Mesnard, Bobak Shahriari, Alexandre Ram \'e , et al. 2024. Gemma 2: Improving open language models at a practical size. arXiv preprint arXiv:2408.00118

    2024 arXiv

  96. [104]

    Erico Tjoa and Cuntai Guan. 2020. A survey on explainable artificial intelligence (xai): Toward medical xai. IEEE transactions on neural networks and learning systems, 32(11):4793--4813

    2020

  97. [105]

    Uno and S

    R. Uno and S. Imaizumi. 2025. https://doi.org/10.31234/osf.io/3qa25 Sensing minimal self in a sentence that involves the speaker . Preprint available at OSF

    2025 doi

  98. [106]

    Camila Valenzuela-Moguillansky and Alejandra V \'a squez-Rosati. 2019. An analysis procedure for the micro-phenomenological interview. Constructivist Foundations, 14(2):123--145

    2019

  99. [107]

    Kush R Varshney and Homa Alemzadeh. 2017. On the safety of machine learning: Cyber-physical systems, decision sciences, and data products. Big data, 5(3):246--255

    2017

  100. [108]

    Vivitha Vijayan and Sharvari Govilkar. 2019. https://doi.org/10.26438/ijcse/v7i3.10651069 A survey on author profiling techniques . International Journal of Computer Sciences and Engineering, 7:1065--1069

    2019 doi

  101. [109]

    Caroline Wang, Bin Han, Bhrij Patel, and Cynthia Rudin. 2023. In pursuit of interpretable, fair and accurate machine learning for criminal recidivism prediction. Journal of Quantitative Criminology, 39(2):519--581

    2023

  102. [110]

    Ling Wang, Jinglin Li, Boyang Zhuang, Shasha Huang, Meilin Fang, Cunze Wang, Wen Li, Mohan Zhang, and Shurong Gong. 2025. Accuracy of large language models when answering clinical research questions: Systematic review and network meta-analysis. Journal of Medical Internet Rese...

    2025

  103. [111]

    Theodore EA Waters and Robyn Fivush. 2015. Relations between narrative coherence, identity, and psychological well-being in emerging adulthood. Journal of personality, 83(4):441--451

    2015

  104. [112]

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems, 35:24824--24837

    2022

  105. [113]

    Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, R \'e mi Louf, Morgan Funtowicz, et al. 2020. Transformers: State-of-the-art natural language processing. In Proceedings of the 2020 conference on empirical met...

    2020

  106. [114]

    Lee Xiong, Chenyan Xiong, Ye Li, Kwok-Fung Tang, Jialin Liu, Paul Bennett, Junaid Ahmed, and Arnold Overwijk. 2020. Approximate nearest neighbor negative contrastive learning for dense text retrieval. arXiv preprint arXiv:2007.00808

    2020 arXiv

  107. [115]

    Han-xue Yang, Han-yu Zhou, Simon SY Lui, and Raymond CK Chan. 2024. Interoception in mental disorders: from self-awareness to interventions. Proceedings of the European Academy of Sciences and Arts, 3

    2024

  108. [116]

    Dan Zahavi. 2008. Subjectivity and selfhood: Investigating the first-person perspective. MIT press

    2008

  109. [117]

    Susan Zhang, Stephen Roller, Naman Goyal, Mikel Artetxe, Moya Chen, Shuohui Chen, Christopher Dewan, Mona Diab, Xian Li, Xi Victoria Lin, et al. 2022. Opt: Open pre-trained transformer language models. arXiv preprint arXiv:2205.01068

    2022 arXiv

  110. [118]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list...

  111. [119]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.