pith. sign in

arxiv: 2605.02832 · v2 · pith:3VPBJN6Pnew · submitted 2026-05-04 · 💻 cs.AI · cs.HC· cs.SE

HAAS: A Policy-Aware Framework for Adaptive Task Allocation Between Humans and Artificial Intelligence Systems

Vicente Pelechano , Antoni Mestre , Manoli Albert , Miriam Gil This is my paper

Pith reviewed 2026-05-20 23:45 UTC · model grok-4.3

classification 💻 cs.AI cs.HCcs.SE
keywords human-AI collaborationtask allocationadaptive governancecontextual banditsautonomy spectrummanufacturingsoftware engineering
0
0 comments X

The pith

Governance constraints in human-AI task allocation act as tunable variables that shift assignments and produce domain-specific gains.

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

The paper presents HAAS as a working system for dividing tasks between humans and AI by treating oversight rules as adjustable settings rather than binary decisions. It pairs a fixed expert system that blocks disallowed AI actions with a learning component that picks the best permitted collaboration style from measured results. Task suitability gets scored on five observable cognitive aspects across a scale of five autonomy levels, tested in both coding and factory environments. A reader would care if this lets groups safely compare different levels of control before locking them in, and if it shows that added rules can sometimes raise output and lower strain together.

Core claim

The central claim is that governance is not a binary switch but a tunable design variable. Tighter constraints predictably convert autonomous AI assignments into supervised collaborations, with domain-specific costs and benefits. In manufacturing, stronger governance can improve operational performance and reduce fatigue simultaneously through a workload-buffering effect. No single governance setting dominates across all contexts; moderate levels grow more competitive as the learner gathers experience inside the allowed action space.

What carries the argument

The two coupled components of the HAAS framework: a rule-based expert system that enforces governance constraints before learning begins, paired with a contextual-bandit learner that chooses among feasible collaboration modes from outcome feedback.

If this is right

  • Tighter governance rules shift more tasks from fully autonomous AI to supervised human-AI modes.
  • In manufacturing, higher governance levels can raise performance metrics while lowering human fatigue at the same time.
  • Moderate governance settings become stronger performers once the learner has accumulated feedback within the constrained action space.
  • Different governance intensities yield different trade-offs depending on the domain and the amount of prior experience.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Organizations could run HAAS-style simulations to compare policy effects in a controlled setting before applying them to live operations.
  • The tunable governance concept may transfer to other mixed-team settings such as medical decision support or logistics coordination.
  • Further work could test whether the same five dimensions predict fit when newer AI models replace the ones used in the current benchmark.

Load-bearing premise

The five auditable cognitive dimensions and the benchmark correctly measure how tasks fit human or AI agents and give reliable signals about real performance and fatigue.

What would settle it

Deploy HAAS in an actual manufacturing line, apply stronger governance settings, and observe no measurable rise in operational performance or drop in reported fatigue.

Figures

Figures reproduced from arXiv: 2605.02832 by Antoni Mestre, Manoli Albert, Miriam Gil, Vicente Pelechano.

Figure 1
Figure 1. Figure 1: Three-layer architecture of HAAS. The governance layer filters feasible collabo￾ration modes before the bandit selects an allocation; execution outcomes then update both reward and human state for the next cycle. (2000), translating qualitative task properties into a continuous AI affinity score. Each subtask s is characterised by a vector d(s) = (r, τ, c, a, h) ∈ [0, 1]5 of rubric-based scores assigned on… view at source ↗
Figure 2
Figure 2. Figure 2: Five-step execution loop connecting governed allocation to the benchmark view at source ↗
Figure 3
Figure 3. Figure 3: Screenshot of the Human–AI Symbiosis Studio dashboard (KPI summary view, manufacturing domain). The panel shows sprint-level KPIs, collaboration-mode distribution, and allocation history charts generated during a benchmark run. current normalised fatigue level (updated per the dynamics in Section 3.6). M(t) ∈ [0, 1] is the monotony signal, defined as the fraction of the last five subtasks assigned to the s… view at source ↗
Figure 4
Figure 4. Figure 4: Governance Ladder — quality, fatigue, lead time, and cumulative regret per level view at source ↗
Figure 5
Figure 5. Figure 5: Collaboration mode redistribution across governance levels. view at source ↗
Figure 6
Figure 6. Figure 6: Best governance level per scenario (portability battery, 10 seeds, 8 cycles). Bar view at source ↗
Figure 7
Figure 7. Figure 7: Long-horizon stability (16 cycles, 30 seeds) for L0, L2, and L4 on four outcome view at source ↗
Figure 8
Figure 8. Figure 8: Multi-dimensional trade-off radar for L0–L4 (standard scenario per domain, 30 view at source ↗
read the original abstract

Deciding how to distribute work between humans and AI systems is a central challenge in organisational design. Most approaches treat this as a binary choice, yet the operational reality is richer: humans and AI routinely share tasks or take complementary roles depending on context, fatigue, and the stakes involved. Governing that distribution -- balancing efficiency, oversight, and human capability -- remains an open problem. This paper presents Human-AI Adaptive Symbiosis (HAAS), an implemented framework for adaptive task allocation in software engineering and manufacturing. HAAS combines two coupled components: a rule-based expert system that enforces governance constraints before any learning occurs, and a contextual-bandit learner that selects among feasible collaboration modes from outcome feedback. Task-agent fit is represented through five auditable cognitive dimensions and a five-mode autonomy spectrum -- from human-only to fully autonomous -- embedded in a reproducible benchmark spanning both domains. Three empirical findings emerge. First, governance is not a binary switch but a tunable design variable: tighter constraints predictably convert autonomous AI assignments into supervised collaborations, with domain-specific costs and benefits. Second, in manufacturing, stronger governance can improve operational performance and reduce fatigue simultaneously -- a workload-buffering effect that contradicts the usual framing of governance as pure overhead. Third, no single governance setting dominates across all contexts; moderate governance becomes increasingly competitive as the learner accumulates experience within the governed action space. Together, these findings position HAAS as a pre-deployment workbench for comparing and inspecting human--AI allocation policies before organisational commitment.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper presents the Human-AI Adaptive Symbiosis (HAAS) framework for adaptive task allocation in software engineering and manufacturing. HAAS couples a rule-based expert system that enforces governance constraints with a contextual-bandit learner that selects among feasible collaboration modes (five-mode autonomy spectrum) using outcome feedback. Task-agent fit is encoded via five auditable cognitive dimensions embedded in a reproducible benchmark. The central claims are three empirical findings: governance acts as a tunable design variable rather than a binary switch; stronger governance in manufacturing can simultaneously improve operational performance and reduce fatigue (workload-buffering effect); and no single governance level dominates, with moderate settings gaining competitiveness as the learner gains experience.

Significance. If the empirical results hold after validation, the work offers a concrete pre-deployment workbench for inspecting human-AI allocation policies, demonstrating that governance constraints can be treated as optimizable parameters with domain-specific trade-offs rather than pure overhead. The reproducible benchmark and the explicit separation of rule-based governance from the learner are positive features that support policy inspection. The significance is limited by the absence of external validation for the cognitive dimensions and fatigue metrics.

major comments (3)
  1. [paragraph on task-agent fit and benchmark] Paragraph on task-agent fit and benchmark: the five auditable cognitive dimensions are introduced as capturing task-agent fit and generating reproducible outcome feedback, yet no inter-rater reliability, external validation against real-world data, or ablation on dimension definitions is reported. This is load-bearing for the second finding, because the workload-buffering effect (improved performance plus reduced fatigue under stronger governance) could be an artifact if the dimensions overweight cognitive load relative to physical ergonomics or if the fatigue metric is derived from the same simulation loop optimized by the learner.
  2. [Abstract] Abstract and empirical findings: the three reported findings are stated without accompanying quantitative results, error bars, sample sizes, statistical tests, or exclusion criteria. Without these details the claims that tighter governance predictably converts autonomous assignments into supervised collaborations and that moderate governance becomes competitive cannot be evaluated for robustness or effect size.
  3. [contextual-bandit learner description] Description of the contextual-bandit learner: the performance gains are produced by the learner operating inside the rule-constrained action space, but the manuscript does not demonstrate that the outcome feedback is independent of the governance rules or the dimension definitions. If feedback is generated internally by the benchmark that encodes the same five dimensions, the reported domain-specific benefits risk circularity rather than constituting an independent empirical result.
minor comments (2)
  1. The five-mode autonomy spectrum is referenced repeatedly but its exact mapping to collaboration modes (e.g., what constitutes 'supervised' versus 'complementary') is not tabulated; adding an explicit table would improve clarity.
  2. Notation for the governance constraint tightness parameter is introduced without a dedicated symbol or equation; consistent use of a single symbol (e.g., G) across sections would reduce ambiguity.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed review. The comments identify valuable opportunities to strengthen the transparency and robustness of the HAAS framework. We respond point-by-point to the major comments below, indicating where revisions will be incorporated.

read point-by-point responses

  1. Referee: Paragraph on task-agent fit and benchmark: the five auditable cognitive dimensions are introduced as capturing task-agent fit and generating reproducible outcome feedback, yet no inter-rater reliability, external validation against real-world data, or ablation on dimension definitions is reported. This is load-bearing for the second finding, because the workload-buffering effect (improved performance plus reduced fatigue under stronger governance) could be an artifact if the dimensions overweight cognitive load relative to physical ergonomics or if the fatigue metric is derived from the same simulation loop optimized by the learner.

    Authors: We agree that further validation would strengthen the workload-buffering claim. The five dimensions are explicitly defined from cognitive load theory and human-factors literature to support auditability and reproducibility inside the controlled benchmark; inter-rater reliability does not apply because the dimensions are not subjectively scored. In the revised manuscript we will add an ablation study that perturbs dimension weights and reports sensitivity of performance and fatigue metrics. We will also expand the limitations section to acknowledge the absence of external real-world validation and outline plans for future domain-expert studies. revision: yes

  2. Referee: Abstract and empirical findings: the three reported findings are stated without accompanying quantitative results, error bars, sample sizes, statistical tests, or exclusion criteria. Without these details the claims that tighter governance predictably converts autonomous assignments into supervised collaborations and that moderate governance becomes competitive cannot be evaluated for robustness or effect size.

    Authors: The abstract follows the conventional format of a high-level summary; all quantitative details, including simulation run counts, statistical tests, effect sizes, and exclusion criteria, appear in the results section. To address the concern we will revise the abstract to include concise quantitative highlights (e.g., observed performance gains and fatigue reductions under moderate governance) while preserving length limits, and we will add explicit cross-references in the main text to the statistical analyses already reported. revision: partial

  3. Referee: Description of the contextual-bandit learner: the performance gains are produced by the learner operating inside the rule-constrained action space, but the manuscript does not demonstrate that the outcome feedback is independent of the governance rules or the dimension definitions. If feedback is generated internally by the benchmark that encodes the same five dimensions, the reported domain-specific benefits risk circularity rather than constituting an independent empirical result.

    Authors: The benchmark supplies fixed outcome models for task performance and fatigue that are independent of any particular governance setting; governance rules only restrict the feasible action space presented to the learner. This separation lets us isolate the effect of different constraint levels on adaptation. In the revision we will insert a clarifying diagram and text that explicitly separates the rule-based governance module from the benchmark feedback loop, and we will add comparative learning-curve experiments across governance regimes to illustrate the non-circular nature of the reported benefits. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation uses external outcome feedback

full rationale

The paper describes HAAS as combining a rule-based expert system enforcing governance constraints with a contextual-bandit learner that selects collaboration modes from outcome feedback in a reproducible benchmark. The key empirical claims (tunable governance, workload-buffering effect in manufacturing, no single setting dominating) are presented as results of the learner operating inside the constrained action space and receiving feedback external to the governance rules themselves. The five auditable cognitive dimensions and autonomy spectrum are introduced as representations for task-agent fit embedded in the benchmark, but the reported performance gains are not shown to reduce by construction to fitted parameters or quantities defined inside the same equations. No self-definitional steps, fitted inputs renamed as predictions, or load-bearing self-citation chains appear in the derivation. The framework is therefore self-contained against the benchmark's external feedback loop.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 2 invented entities

The framework rests on the assumption that outcome feedback from the benchmark is sufficient to train the bandit and that the five cognitive dimensions are stable and auditable; no explicit free parameters are named, but the governance tightness acts as a tunable knob.

free parameters (1)
  • governance constraint tightness
    Tunable parameter in the expert system that converts autonomous assignments into supervised ones; its value is chosen per domain and affects reported performance.
axioms (1)
  • domain assumption Contextual bandits can improve allocation decisions from repeated outcome feedback within the feasible action space defined by governance rules.
    Invoked in the description of the learner component that selects among collaboration modes.
invented entities (2)
  • Five auditable cognitive dimensions for task-agent fit no independent evidence
    purpose: Represent collaboration requirements so that allocation decisions can be inspected and governed.
    New modeling construct introduced to make human-AI matching explicit and auditable.
  • Five-mode autonomy spectrum no independent evidence
    purpose: Define discrete collaboration modes from human-only to fully autonomous AI.
    Core representational device that structures the action space for both the expert system and the learner.

pith-pipeline@v0.9.0 · 5809 in / 1550 out tokens · 57138 ms · 2026-05-20T23:45:29.755828+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

60 extracted references · 60 canonical work pages · 1 internal anchor

  1. [1]

    International Journal of Human-Computer Studies , volume =

    Gil, Miriam and Albert, Manoli and Fons, Joan and Pelechano, Vicente , title =. International Journal of Human-Computer Studies , volume =. 2019 , doi =

    work page 2019

  2. [2]

    and Wickens, Christopher D

    Parasuraman, Raja and Sheridan, Thomas B. and Wickens, Christopher D. , title =. IEEE Transactions on Systems, Man, and Cybernetics -- Part A: Systems and Humans , year =

    work page

  3. [3]

    , title =

    Sheridan, Thomas B. , title =

    work page

  4. [4]

    , title =

    Cummings, Mary L. , title =. IEEE Intelligent Systems , year =

    work page

  5. [5]

    and Demir, Mustafa and Cooke, Nancy J

    McNeese, Nathan J. and Demir, Mustafa and Cooke, Nancy J. and Myers, Christopher , title =. Human Factors , year =

    work page

  6. [6]

    and Robert, Lionel P

    Ali, Arsha and Azevedo-Sa, Hebert and Tilbury, Dawn M. and Robert, Lionel P. , title =. Scientific Reports , year =

    work page

  7. [7]

    Procedia CIRP , year =

    Petzoldt, Christoph and Niermann, Dario and Keiser, Dennis and Freitag, Michael , title =. Procedia CIRP , year =

    work page

  8. [8]

    Systems , year =

    Urrea, Claudio , title =. Systems , year =

    work page

  9. [9]

    Artificial Intelligence for Engineering Design, Analysis and Manufacturing , year =

    Kirgil-Budakli, Rukiye and Zeng, Yong and Akgunduz, Ali , title =. Artificial Intelligence for Engineering Design, Analysis and Manufacturing , year =

    work page

  10. [10]

    and de Vreede, Gert-Jan and de Vreede, Triparna and Elkins, Aaron and Maier, Ronald and Merz, Alexander B

    Seeber, Isabella and Bittner, Eva and Briggs, Robert O. and de Vreede, Gert-Jan and de Vreede, Triparna and Elkins, Aaron and Maier, Ronald and Merz, Alexander B. and Oeste-Reiß, Sarah and Randrup, Nils and Schwabe, Gerhard and Söllner, Matthias , title =. Information & Management , year =

    work page

  11. [11]

    The International Journal of Robotics Research , year =

    Soh, Harold and Xie, Yicheng and Chen, Min and Halpern, David , title =. The International Journal of Robotics Research , year =

    work page

  12. [12]

    and Billings, Dee R

    Hancock, Peter A. and Billings, Dee R. and Schaefer, Kristin E. and Chen, Jessie Y. C. and de Visser, Ewart J. and Parasuraman, Raja , title =. Human Factors , year =

    work page

  13. [13]

    Hybrid Intelligence , journal =

    Dellermann, Dominik and Ebel, Philipp and S. Hybrid Intelligence , journal =. 2019 , volume =

    work page 2019

  14. [14]

    2020 , url =

    Lattimore, Tor and Szepesvári, Csaba , title =. 2020 , url =

    work page 2020

  15. [15]

    Finite-time analysis of the multiarmed bandit problem , journal =

    Auer, Peter and Cesa-Bianchi, Nicol\`. Finite-time analysis of the multiarmed bandit problem , journal =. 2002 , volume =

    work page 2002

  16. [16]

    , title =

    Li, Lihong and Chu, Wei and Langford, John and Schapire, Robert E. , title =. Proceedings of the 19th International Conference on World Wide Web (WWW) , year =

    work page

  17. [17]

    and Van Roy, Benjamin and Kazerouni, Abbas and Osband, Ian and Wen, Zheng , title =

    Russo, Daniel J. and Van Roy, Benjamin and Kazerouni, Abbas and Osband, Ian and Wen, Zheng , title =. Foundations and Trends in Machine Learning , year =

    work page

  18. [18]

    European Conference on Machine Learning (ECML) , year =

    Kocsis, Levente and Szepesvári, Csaba , title =. European Conference on Machine Learning (ECML) , year =

    work page

  19. [19]

    Algorithmic Learning Theory (ALT) , year =

    Garivier, Aurélien and Moulines, Eric , title =. Algorithmic Learning Theory (ALT) , year =

    work page

  20. [20]

    Monarch, Robert , title =

    work page

  21. [21]

    International Journal of Human-Computer Interaction , year =

    Shneiderman, Ben , title =. International Journal of Human-Computer Interaction , year =

    work page

  22. [22]

    2024 , type =

    Regulation (EU) 2024/1689 of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act) , institution =. 2024 , type =

    work page 2024

  23. [23]

    2016 , number =

    Robots and robotic devices -- Collaborative robots , institution =. 2016 , number =

    work page 2016

  24. [24]

    2018 , url =

    Dafoe, Allan , title =. 2018 , url =

    work page 2018

  25. [25]

    and Dodson, John D

    Yerkes, Robert M. and Dodson, John D. , title =. Journal of Comparative Neurology and Psychology , year =

    work page

  26. [26]

    , title =

    Parasuraman, Raja and Wickens, Christopher D. , title =. Human Factors , year =

    work page

  27. [27]

    Automatica , year =

    Bainbridge, Lisanne , title =. Automatica , year =

    work page

  28. [28]

    , title =

    Endsley, Mica R. , title =. Human Factors , year =

    work page

  29. [29]

    , title =

    Crandall, Beth and Klein, Gary and Hoffman, Robert R. , title =

    work page

  30. [30]

    2023 , doi =

    Brynjolfsson, Erik and Li, Danielle and Raymond, Lindsey , title =. 2023 , doi =

    work page 2023

  31. [31]

    and Van Mieghem, Jan A

    Gijsbrechts, Joren and Boute, Robert N. and Van Mieghem, Jan A. and Zhang, Dennis J. , title =. Manufacturing & Service Operations Management , year =

    work page

  32. [32]

    Machine Learning , year =

    Ben-David, Shai and Blitzer, John and Crammer, Koby and Kulesza, Alex and Pereira, Fernando and Vaughan, Jennifer , title =. Machine Learning , year =

    work page

  33. [33]

    Science , year =

    Noy, Shakked and Zhang, Whitney , title =. Science , year =

    work page

  34. [34]

    and Lifshitz-Assaf, Hila and Kellogg, Katherine C

    Dell'Acqua, Fabrizio and McFowland, Edward and Mollick, Ethan R. and Lifshitz-Assaf, Hila and Kellogg, Katherine C. and Rajendran, Saran and Krayer, Lisa and Candelon, Fran. Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of. 2023 , number =. doi:10.2139/ssrn.4573321 , url =

    work page doi:10.2139/ssrn.4573321 2023

  35. [35]

    The Impact of AI on Developer Productivity: Evidence from GitHub Copilot

    Peng, Sida and Kalliamvakou, Eirini and Cihon, Peter and Demirer, Mert , title =. 2023 , howpublished =. doi:10.48550/arXiv.2302.06590 , url =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2302.06590 2023

  36. [36]

    and See, Katrina A

    Lee, John D. and See, Katrina A. , title =. Human Factors , year =

    work page

  37. [37]

    Artificial Intelligence , year =

    Miller, Tim , title =. Artificial Intelligence , year =

    work page

  38. [38]

    To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on

    Bu. To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on. Proceedings of the. 2021 , volume =

    work page 2021

  39. [39]

    Proceedings of the 25th International Joint Conference on Artificial Intelligence (

    Kamar, Ece , title =. Proceedings of the 25th International Joint Conference on Artificial Intelligence (. 2016 , pages =

    work page 2016

  40. [40]

    Proceedings of the 2021

    Bansal, Gagan and Wu, Tongshuang and Zhou, Joyce and Fok, Raymond and Nushi, Besmira and Kamar, Ece and Ribeiro, Marco Tulio and Weld, Daniel , title =. Proceedings of the 2021. 2021 , pages =

    work page 2021

  41. [41]

    and Parasuraman, Raja and Matthews, Gerald , title =

    Warm, Joel S. and Parasuraman, Raja and Matthews, Gerald , title =. Human Factors , year =

    work page

  42. [42]

    , title =

    Parasuraman, Raja and Manzey, Dietrich H. , title =. Human Factors , year =

    work page

  43. [43]

    and Fisk, Arthur D

    Beer, Jenay M. and Fisk, Arthur D. and Rogers, Wendy A. , title =. Journal of Human-Robot Interaction , year =

    work page

  44. [44]

    2011 , number =

    Robots and robotic devices -- Safety requirements for industrial robots -- Part 1: Robots , institution =. 2011 , number =

    work page 2011

  45. [45]

    , title =

    Fitts, Paul M. , title =. 1951 , note =

    work page 1951

  46. [46]

    and Kaber, David B

    Endsley, Mica R. and Kaber, David B. , title =. Ergonomics , year =

    work page

  47. [47]

    and Endsley, Mica R

    Kaber, David B. and Endsley, Mica R. , title =. Theoretical Issues in Ergonomics Science , year =

    work page

  48. [48]

    Handbook of Cognitive Task Design , editor =

    Inagaki, Toshiyuki , title =. Handbook of Cognitive Task Design , editor =. 2003 , pages =

    work page 2003

  49. [49]

    Expert Systems with Applications , year =

    Liao, Shu-Hsien , title =. Expert Systems with Applications , year =

    work page

  50. [50]

    Turban, Efraim , title =

    work page

  51. [51]

    and Riley, Gary D

    Giarratano, Joseph C. and Riley, Gary D. , title =

    work page

  52. [52]

    Academy of Management Review , year =

    Raisch, Sebastian and Krakowski, Sebastian , title =. Academy of Management Review , year =

    work page

  53. [53]

    and Inkpen, Kori and Teevan, Jaime and Kiber, Ruth and Horvitz, Eric , title =

    Amershi, Saleema and Weld, Dan and Vorvoreanu, Mihaela and Fourney, Adam and Nushi, Besmira and Collisson, Penny and Suh, Jina and Iqbal, Shamsi and Bennett, Paul N. and Inkpen, Kori and Teevan, Jaime and Kiber, Ruth and Horvitz, Eric , title =. Proceedings of the 2019. 2019 , pages =

    work page 2019

  54. [54]

    Hemmer, Patrick and Schemmer, Max and V. Human-. 2021 , howpublished =

    work page 2021

  55. [55]

    2022 , doi =

    Shneiderman, Ben , title =. 2022 , doi =

    work page 2022

  56. [56]

    Minds and Machines , year =

    Floridi, Luciano and Cowls, Josh and Beltrametti, Monica and Chatila, Raja and Chazerand, Patrice and Dignum, Virginia and Luetge, Christoph and Madelin, Robert and Pagallo, Ugo and Rossi, Francesca and Schafer, Burkhard and Valcke, Peggy and Vayena, Effy , title =. Minds and Machines , year =

    work page

  57. [57]

    Nature Human Behaviour , year =

    Vaccaro, Michelle and Almaatouq, Abdullah and Malone, Thomas , title =. Nature Human Behaviour , year =

    work page

  58. [58]

    and Flathmann, Christopher and McNeese, Nathan J

    Hauptman, Allyson I. and Flathmann, Christopher and McNeese, Nathan J. , title =. Applied Ergonomics , year =

    work page

  59. [59]

    and Heidari, Hoda and Jalali, Mohammad S

    Gonzalez, Cleotilde and Donahue, Kate and Goldstein, Daniel G. and Heidari, Hoda and Jalali, Mohammad S. and Schelble, Beau and Singh, Aarti and Woolley, Anita Williams , title =. PNAS Nexus , year =

    work page

  60. [60]

    Advanced Engineering Informatics , year =

    Wang, Jingfei and Yan, Yan and Hu, Yaoguang and Yang, Xiaonan , title =. Advanced Engineering Informatics , year =

    work page