Pith. sign in

REVIEW 4 major objections 6 minor 52 references

AnyAni: An Interactive System with Generative AI for Animation Effect Creation and Code Understanding in Web Development

T0 review · 4 major / 6 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 AnyAni is a human-AI collaborative system that helps front-end developers create, iterate, and understand web animation effects, and its nine-participant evaluation reports significantly better perceived performance than a WebStorm plus…

desk verdict AnyAni is a thoughtful systems contribution with a plausible but unproven usability claim; the evaluation is too thin to support the 'significantly enhanced performance' headline. read the letter →

arxiv 2506.21962 v1 pith:TGRI6244 submitted 2025-06-27 cs.HC

Tianrun Qiu , Yuxin Ma This is my paper
classification cs.HC
keywords creativitysupportwebanimationgenerativeAIlargelanguagemodelshuman-AIcollaborationcodeunderstandingversioncontroluserstudy
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 targets a gap between code generation and visual design: front-end developers who are not trained designers struggle to describe, design, and implement web animations, and ordinary LLM chat interfaces add to that struggle by returning long, lazy, or unmodifiable code. AnyAni is the authors' proposed answer: a web-based human-AI system that combines structured incremental code generation, a tree-shaped version control panel for nonlinear exploration, and interactive code-understanding tools such as semantic diffs and live parameter editing. The paper seeks to establish that this workflow makes animation creation and code comprehension easier for developers, supporting the claim with a formative interview study (N=6), a user study (N=9), and a baseline comparison against WebStorm with GPT-4o. The reported quantitative support includes an overall PSSUQ usability score of 6.11/7 and a NASA-TLX performance score significantly better than baseline (p=0.0013). If the central claim holds, it suggests a general recipe for turning raw LLM code generation into a structured, learnable, iterative creative practice.

What carries the argument

The load-bearing mechanism is incremental generation with line-number-marked code. Before each LLM call, code is beautified and each line prefixed with '(No.) |'; the model returns structured insert/remove/update operations referencing those line numbers, which are applied to the existing code and then cleaned. This preserves untouched sections across iterations, prevents the model from 'lazily' omitting code, and keeps style consistent. Around this core, the system adds a tree-based version control panel for nonlinear exploration, AST-based semantic diff explanations, interactive parameter manipulation, and a video-understanding loop that checks rendered animations against requirements.

What would settle it

Run the same two animation tasks with a larger, independent sample in a between-subjects design, an interviewer who did not build the system, and objective metrics such as task completion time, number of manual edits, and expert-rated output quality. If AnyAni does not beat WebStorm plus GPT-4o on those objective measures, the paper's central performance claim is refuted.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the barriers to AI-assisted animation creation are not mainly model capability but workflow structure. The authors show that GPT-4o can generate syntactically valid CSS animation code, yet raw chat interaction fails developers through overlong output, omitted 'lazy' code, style drift across iterations, and opaque diffs. AnyAni's contribution is a workflow that compensates for these failures: code is reformatted with line numbers and modified through explicit insert/remove/update operations so long code bases survive iteration; a version tree preserves and isolates branches; and the chat panel is augmented with AST-based semantic diff summaries, parameter sliders, and automatic video-based visual checks. The resulting usability evidence is the paper's central empirical claim, with performance measured by NASA-TLX reported as significantly better than the baseline.

Load-bearing premise

The central claim rests on self-reported ratings from nine developers recruited from the authors' professional networks and interviewed by the person who built the system, with no objective measures of speed or quality; if those ratings reflect novelty or politeness rather than real improvement, the usability result collapses.

Editorial extensions

If this is right

  • Developers without design training can take a vague idea like 'a cool rotating effect' and end with runnable CSS animation code, because visual scaffolding and prompt templates translate the idea into concrete specifications.
  • Long existing code can be modified iteratively by an LLM without being rewritten, since line-numbered incremental edits leave untouched lines intact.
  • Users can learn from generated code through AST-based semantic diffs and live parameter editing, turning the AI output into a practice-enabled tutorial rather than a black box.
  • Automatic visual correction closes the loop between what the user asked for and what actually renders, covering aspects of quality that syntactic code repair cannot address.
  • The nonlinear version tree supports parallel exploration of design directions, so users can compare and merge alternatives without losing earlier states.

Reading between the lines

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

  • If the workflow pattern holds, the same combination of structured incremental edits, version trees, and interactive parameter controls could extend to other creative coding domains, such as data visualization, generative UI design, or game prototyping, where iteration and code opacity are similar bottlenecks.
  • A testable extension the paper leaves implicit is an independent evaluation: because the quantitative evidence is entirely self-reported and collected by the system's builder, an objective study comparing completion time, edit counts, and expert-rated output quality would be needed before treating the performance advantage as settled.
  • The video-understanding feedback loop suggests a general template for verifying generative visual output: record the rendered artifact, have a multimodal model compare it against the specification, and feed the unmet parts back into the generator.
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

4 major / 6 minor

Summary. The paper presents AnyAni, a web-based human-AI collaborative system that assists front-end developers in creating, iterating on, and understanding CSS animation effects. The work includes a formative interview study with six developers to identify design challenges, a system prototype with structured/incremental LLM generation, a nonlinear version-control panel, code-understanding aids, and video-based automatic correction. The authors then report a user study with nine developers comparing AnyAni against a WebStorm plus GPT-4o baseline, using NASA-TLX and PSSUQ questionnaires plus qualitative interviews and an interaction-log heatmap. The abstract and introduction claim that AnyAni 'demonstrated usability' and 'significantly enhanced users' performance,' based primarily on one NASA-TLX Performance subscale result (p=0.0013).

Significance. If the claims are supported, AnyAni would be a useful contribution to AI-assisted creative coding: the incremental line-number-based generation protocol, the node-based version tree, the semantic diff and code-explanation features, and the video-understanding-based correction loop are concrete and interesting system mechanisms. The formative study is coherent and motivates the design goals reasonably. The N=9 user study is small but not unusual for a usability investigation in this subfield. However, the paper's central quantitative claim of significantly enhanced performance is not yet established by the evidence as presented; the main value at this stage is the system design and the qualitative findings, which need to be separated from the stronger inferential language.

major comments (4)
  1. [6.2.2] The abstract and Section 1 claim that AnyAni 'significantly enhanced users' performance,' but the only inferential support is the NASA-TLX Performance subscale (p=0.0013, N=9). The paper does not report a multiple-comparison correction across the six TLX subscales, and only this one subscale reaches significance; Frustration is reported at p=0.0622 and is not significant. The Section 1 claim of an advantage 'in multiple dimensions' is therefore not supported by the statistics as presented. Please report all subscale tests with an appropriate correction, or weaken the claim to the specific subscale that was significant.
  2. [6.2.4] The performance claim is not corroborated by objective measures. Section 6.2.4 states that click data and usage statistics were automatically collected, but the analysis is only a qualitative heatmap; no counts, durations, or inferential tests are reported, and no objective measure of task completion, code correctness, or final animation quality is provided anywhere in Section 6. As a result, the statement that AnyAni 'significantly enhanced users' performance' rests entirely on unverified self-report.
  3. [6.2.1] The text says participants completed the PSSUQ 'for both our system and the baseline (WebStorm + GPT-4o),' but only the AnyAni overall score (6.11/7) is reported; no baseline PSSUQ results or comparison test are given. Please provide the baseline PSSUQ data or explicitly state that PSSUQ was collected only for AnyAni.
  4. [6.1] The evaluation sessions were organized and conducted one-on-one by the first author, who also built AnyAni, and participants were recruited from the authors' professional networks and compensated. This design creates a real risk of demand characteristics and social desirability bias in all self-report measures. At minimum, this should be acknowledged as a threat to validity in Section 7; ideally the study would use a neutral facilitator or blinded protocol, but as written the paper does not discuss this limitation.
minor comments (6)
  1. [5.3] There are typos in Section 5.3: 'especcially' should be 'especially' and 'comparation' should be 'comparison.'
  2. [3.2.3] In Section 3.2.3, 'owing the the complexity' should be 'owing to the complexity.'
  3. [6.2.2] The sentence 'it is evident that our system visibly outperforms the baseline across four indicators' is not supported by the reported statistics, since only the Performance subscale is significant; mental and physical demand are said to be similar. Please rephrase to distinguish descriptive differences from statistically supported ones.
  4. [6.2.2] The NASA-TLX scoring description is confusing: the text says '0 indicates complete success and 20 indicates complete failure,' but this is not the standard TLX Performance anchor wording. Please clarify the exact anchors used.
  5. [A] Table 1 has a duplicated 'Experience' header, and the legend for NOV/MID/EXP is placed after the table; moving the legend before the table would improve readability.
  6. [6.2.3] In Section 6.2.3, 'participants are also pleasant about the subsequent interaction process' should be 'participants were also pleased with the subsequent interaction process.'

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's claims are empirical usability findings, not derivations from their own inputs.

full rationale

The paper contains no mathematical derivation, fitted parameter, or first-principles prediction whose output is equivalent to its input by construction. The central claims—that AnyAni supports animation creation, code understanding, and enhanced user performance—are empirical usability claims evaluated through a user study (Sections 6.1 and 6.2), supported by NASA-TLX, PSSUQ, participant interviews, and interaction-log heat maps. The highlighted inferential result, the NASA-TLX performance comparison (AnyAni 4.44 vs baseline 11.5, p=0.0013, N=9), is a between-system comparison of self-reported ratings, not a quantity defined in terms of the system's design or fitted to the same data it is said to predict. No load-bearing self-citation appears in the argument: the cited prior systems and methods are external works, and the authors' own system is evaluated rather than cited as authoritative support. The limitation statements in Sections 7.1 and 7.2 acknowledge scope constraints, but these concern generalizability and evaluation design, not circular reasoning. Concerns about small sample size, recruitment from professional networks, administration by the first author, and reliance on self-report are validity and bias risks, not circularity: the statistical claim is not forced by construction or by a self-citation chain. Therefore no circularity is identified, and the appropriate score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the formative study (N=6) surfacing generalizable pain points, on a nine-person user study accurately measuring real-world usability, and on the stability of proprietary LLM services. No numeric parameters are fitted to data. The main load-bearing premises are qualitative assumptions about participant representativeness and measurement validity.

assumptions (5)
  • domain assumption Six semi-structured interviews with front-end developers who have six or more years of experience surface the main challenges of the broader target population.
    Section 3.1 reports N=6; no saturation analysis is reported.
  • domain assumption Nine evaluation participants with at least two years of front-end development experience adequately represent developers who need animation support.
    Section 6.1 and Table 1; the sample is small and recruited via the authors' networks.
  • domain assumption Self-reported NASA-TLX and PSSUQ scores are valid proxies for actual usability, workload, and creative performance.
    Section 6.2; no objective task output metrics are reported.
  • domain assumption WebStorm plus GPT-4o is a fair baseline that isolates AnyAni's workflow contributions.
    Section 6.1; the baseline lacks preview, version tree, incremental patching, and parameter widgets.
  • domain assumption The observed behavior of GPT-4o and Qwen2.5-VL at study time is stable and representative of the general model capabilities claimed.
    Section 5.5; these are proprietary, versioned services whose responses are not pinned to a specific model snapshot.

how reviews work

0 comments
Cite this review

Pith. "Pith review of AnyAni: An Interactive System with Generative AI for Animation Effect Creation and Code Understanding in Web Development." pith.science (2026). https://pith.science/paper/TGRI6244

@misc{pith2026250621962,
  author       = {Pith},
  title        = {Pith review of: AnyAni: An Interactive System with Generative AI for Animation Effect Creation and Code Understanding in Web Development},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TGRI6244}},
  note         = {Machine review of arXiv:2506.21962}
}
read the original abstract

Generative AI assistants have been widely used in front-end programming. However, besides code writing, developers often encounter the need to generate animation effects. As novices in creative design without the assistance of professional designers, developers typically face difficulties in describing, designing, and implementing desired animations. To address this issue, we conducted a formative study (N=6) to identify the challenges that code developers face when dealing with animation design issues. Then, we introduce AnyAni, a human-AI collaborative system that supports front-end developers in the ideation, manipulation, and implementation of animation effects. The system combines the assistance of generative AI in creative design by adopting a nonlinear workflow for iterative animation development. In addition, developers can understand and learn the code generated for implementing animations through various interactive methods. A user study (N=9) demonstrated the usability of AnyAni in animation effect creation support for developers.

Figures

Figures reproduced from arXiv: 2506.21962 by the authors.

Figure 1
Figure 1. We introduce AnyAni, a human-AI collaborative system that leverage large language models to empower users to [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of AnyAni’s user interface, which consists [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Example of non-linear exploration on the task of [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Example of iterative optimization on a submarine [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Workflow of AnyAni system discrete input-output interaction round (either AI-generated or manually modified). Built upon Vue-Flow’s framework, the system enables intuitive node manipulation through drag-and-drop functionality, with each node embedded in an iframe for v…
Figure 8
Figure 8. Figure 8: Compar￾ing code differ￾ence [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 11
Figure 11. Figure 11: Participants’ response to PSSUQ questionnaire [PITH_FULL_IMAGE:figures/full_fig_p009_11.png]
Figure 12
Figure 12. Figure 12: Comparison of NASA-TLX Cognitive Load Ques [PITH_FULL_IMAGE:figures/full_fig_p009_12.png]
Figure 13
Figure 13. Figure 13: Usage record heat map (blue indicating more clicks [PITH_FULL_IMAGE:figures/full_fig_p010_13.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

52 extracted references · 37 canonical work pages

  1. [1]

    Airbnb. 2024. Lottie Docs. https://airbnb.io/lottie/# [Online; accessed 12 September 2024]

    work page 2024

  2. [2]

    Tyler Angert, Miroslav Suzara, Jenny Han, Christopher Pondoc, and Hariharan Subramonyam. 2023. Spellburst: A node-based interface for exploratory creative coding with natural language prompts. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology (UIST ’23) . Association for Computing Machinery, New York, NY, USA, 1–22

    work page 2023

  3. [3]

    Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, Humen Zhong, Yuanzhi Zhu, Mingkun Yang, Zhaohai Li, Jianqiang Wan, Pengfei Wang, Wei Ding, Zheren Fu, Yiheng Xu, Jiabo Ye, Xi Zhang, Tianbao Xie, Zesen Cheng, Hang Zhang, Zhibo Yang, Haiyang Xu, and Junyang Lin. 2025. Qwen2.5-VL Technical Rep...

    arXiv 2025

  4. [4]

    Ali Borji. 2022. Generated Faces in the Wild: Quantitative Comparison of Stable Diffusion, Midjourney and DALL-E 2. arXiv preprint (2022). arXiv:2210.00586

    arXiv 2022

  5. [5]

    Tim Brooks, Bill Peebles, Connor Holmes, Will DePue, Yufei Guo, Li Jing, David Schnurr, Joe Taylor, Troy Luhman, Eric Luhman, Clarence Ng, Ricky Wang, and Aditya Ramesh. 2024. Video generation models as world simulators. OpenAI Blog. https://openai.com/research/video-generation-models-as-world-simulators

    work page 2024

  6. [6]

    Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin...

    work page 2020

  7. [7]

    E, Zhutian Chen, and Haijun Xia

    Yining Cao, Jane L. E, Zhutian Chen, and Haijun Xia. 2023. DataParticles: Block- Based and Language-Oriented Authoring of Animated Unit Visualizations. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (CHI ’23). Association for Computing Machinery, New York, NY, USA, 1–15

    work page 2023

  8. [8]

    Bay-Wei Chang and David Ungar. 1993. Animation: from cartoons to the user interface. In Proceedings of the 6th Annual ACM Symposium on User Interface Software and Technology (UIST ’93) . 45–55

    work page 1993

Show all 52 references
  1. [9]

    Shoufa Chen, Chongjian Ge, Yuqi Zhang, Yida Zhang, Fengda Zhu, Hao Yang, Hongxiang Hao, Hui Wu, Zhichao Lai, Yifei Hu, Ting-Che Lin, Shilong Zhang, Fu Li, Chuan Li, Xing Wang, Yanghua Peng, Peize Sun, Ping Luo, Yi Jiang, Zehuan Yuan, Bingyue Peng, and Xiaobing Liu. 2025. Goku:...

    2025 arXiv

  2. [10]

    Ruijia Cheng, Titus Barik, Alan Leung, Fred Hohman, and Jeffrey Nichols. 2024. BISCUIT: Scaffolding LLM-Generated Code with Ephemeral UIs in Computational Notebooks. arXiv preprint (2024). arXiv:2404.07387

    2024 arXiv

  3. [11]

    Li-Yuan Chiou, Peng-Kai Hung, Rung-Huei Liang, and Chun-Teng Wang. 2023. Designing with AI: An Exploration of Co-Ideation with Image Generators. In Proceedings of the 2023 ACM Designing Interactive Systems Conference. 1941–1954

    2023

  4. [12]

    Yingchaojie Feng, Xingbo Wang, Kam Kwai Wong, Sijia Wang, Yuhong Lu, Min- feng Zhu, Baicheng Wang, and Wei Chen. 2024. PromptMagician: Interactive Prompt Engineering for Text-to-Image Creation. IEEE Transactions on Visualiza- tion and Computer Graphics 30, 1 (2024), 295–305

    2024

  5. [13]

    Miao Hao and Taile Ni. 2021. Digital Model of Computer-Aided Art Design: From Plane to Space. Computer-Aided Design and Applications 19 (2021), 130–141

    2021

  6. [14]

    Sandra G. Hart. 2006. NASA-Task Load Index (NASA-TLX); 20 Years Later. Proceedings of the Human Factors and Ergonomics Society Annual Meeting 50, 9 (2006), 904–908

    2006

  7. [15]

    Wenyi Hong, Weihan Wang, Ming Ding, Wenmeng Yu, Qingsong Lv, Yan Wang, Yean Cheng, Shiyu Huang, Junhui Ji, Zhao Xue, Lei Zhao, Zhuoyi Yang, Xi- aotao Gu, Xiaohan Zhang, Guanyu Feng, Da Yin, Zihan Wang, Ji Qi, Xixuan Song, Peng Zhang, Debing Liu, Bin Xu, Juanzi Li, Yuxiao Dong,...

  8. [16]

    J. D. Hunter. 2007. Matplotlib: A 2D Graphics Environment.Computing in Science & Engineering 9, 3 (2007), 90–95. https://doi.org/10.1109/MCSE.2007.55

    2007 doi

  9. [17]

    Martin Jonsson and Jakob Tholander. 2022. Cracking the code: Co-coding with AI in creative programming education. In ACM Conferences. Association for Computing Machinery, New York, NY, USA, 5–14. https://doi.org/10.1145/ 3527927.3532801

    2022

  10. [18]

    Anna Kantosalo and Hannu Toivonen. 2016. Modes for Creative Human- Computer Collaboration: Alternating and Task-Divided Co-Creativity. InProceed- ings of the International Conference on Innovative Computing and Cloud Computing (ICCC 2016). Sony CSL Paris, France

    2016

  11. [19]

    Ericson, David Weintrop, and Tovi Grossman

    Majeed Kazemitabaar, Justin Chow, Carl Ka To Ma, Barbara J. Ericson, David Weintrop, and Tovi Grossman. 2023. Studying the effect of AI Code Generators on Supporting Novice Learners in Introductory Programming. In Proceedings of the 2023 CHI Conference on Human Factors in Comp...

    2023

  12. [20]

    Majeed Kazemitabaar, Runlong Ye, Xiaoning Wang, Austin Zachary Henley, Paul Denny, Michelle Craig, and Tovi Grossman. 2024. CodeAid: Evaluating a Classroom Deployment of an LLM-Based Programming Assistant That Balances Student and Educator Needs. In Proceedings of the CHI Conf...

    2024

  13. [21]

    Xingyu Lan, Yang Shi, Yanqiu Wu, Xiaohan Jiao, and Nan Cao. 2021. Kineticharts: Augmenting Affective Expressiveness of Charts in Data Stories with Animation Design. IEEE Transactions on Visualization and Computer Graphics 28, 1 (2021), 933–943

    2021

  14. [22]

    James Lewis. 1992. Psychometric Evaluation of the Post-Study System Usability Questionnaire: The PSSUQ, In Proceedings of the Human Factors Society 36th Annual Meeting. Proceedings of the Human Factors Society 2, 1259–1263

    1992

  15. [23]

    James R. Lewis. 2002. Psychometric Evaluation of the PSSUQ Using Data from Five Years of Usability Studies. International Journal of Human–Computer Interaction 14, 3–4 (2002), 463–488

    2002

  16. [24]

    Liang, Chenyang Yang, and Brad A

    Jenny T. Liang, Chenyang Yang, and Brad A. Myers. 2024. A Large-Scale Survey on the Usability of AI Programming Assistants: Successes and Challenges. In Proceedings of the IEEE/ACM 46th International Conference on Software Engineering (Lisbon, Portugal) (ICSE ’24). Association...

    2024

  17. [25]

    Gaojie Lin, Jianwen Jiang, Jiaqi Yang, Zerong Zheng, and Chao Liang. 2025. OmniHuman-1: Rethinking the Scaling-Up of One-Stage Conditioned Human Animation Models. arXiv:2502.01061 [cs.CV]

    2025 arXiv

  18. [26]

    Vivian Liu and Lydia B. Chilton. 2022. Design Guidelines for Prompt Engineering Text-to-Image Generative Models. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems (CHI ’22) . Association for Computing Machinery, New York, NY, USA, 1–23

    2022

  19. [27]

    Vivian Liu, Rubaiat Habib Kazi, Li-Yi Wei, Matthew Fisher, Timothy Langlois, Seth Walker, and Lydia Chilton. 2024. LogoMotion: Visually Grounded Code Gen- eration for Content-Aware Animation. arXiv preprint (2024). arXiv:2405.07065

    2024 arXiv

  20. [28]

    Hiroaki Mikami, Daisuke Sakamoto, and Takeo Igarashi. 2017. Micro-Versioning Tool to Support Experimentation in Exploratory Programming. In Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems (CHI ’17) . Association for Computing Machinery, New York, N...

    2017

  21. [29]

    Daye Nam, Andrew Macvean, Vincent Hellendoorn, Bogdan Vasilescu, and Brad Myers. 2024. Using an LLM to Help with Code Understanding. In Proceedings of the IEEE/ACM 46th International Conference on Software Engineering . 1–13

    2024

  22. [30]

    Molly Jane Nicholas, Sarah Sterman, and Eric Paulos. 2022. Creative and Motiva- tional Strategies Used by Expert Creative Practitioners. In Proceedings of the 14th Conference on Creativity and Cognition . 323–335

    2022

  23. [31]

    Bin Ning, Fang Liu, and Zhixiong Liu. 2023. Creativity Support in AI Co-Creative Tools: Current Research, Challenges and Opportunities. In2023 26th International Conference on Computer Supported Cooperative Work in Design (CSCWD) . IEEE, 5–10. https://doi.org/10.1109/CSCWD5746...

    2023

  24. [32]

    OpenAI. 2023. GPT-4V(ision) System Card. https://cdn.openai.com/papers/ GPTV_System_Card.pdf [Online]

    2023

  25. [33]

    OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Alt- man, et al. 2023. GPT-4 Technical Report. arXiv preprint (2023). arXiv:2303.08774

    2023 arXiv

  26. [34]

    Fred Paas, Pascal W. M. Van Gerven, and Pieter Wouters. 2007. Instructional Efficiency of Animation: Effects of Interactivity Through Mental Reconstruction of Static Key Frames. Applied Cognitive Psychology: The Official Journal of the Society for Applied Research in Memory an...

    2007

  27. [35]

    Susan Palmiter and Jay Elkerton. 1991. An evaluation of animated demonstrations of learning computer-based tasks. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’91) . Association for Computing Machinery, New York, NY, USA, 257–263

    1991

  28. [36]

    It’s Weird That it Knows What I Want

    James Prather, Brent N. Reeves, Paul Denny, Brett A. Becker, Juho Leinonen, Andrew Luxton-Reilly, Garrett Powell, James Finnie-Ansley, and Eddie Antonio Santos. 2023. “It’s Weird That it Knows What I Want”: Usability and Interactions with Copilot for Novice Programmers. ACM Tr...

    2023

  29. [37]

    Tingting Qiao, Jing Zhang, Duanqing Xu, and Dacheng Tao. 2019. Mirror- gan: Learning Text-to-Image Generation by Redescription. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 1505–1514

    2019

  30. [38]

    Aditya Ramesh, Mikhail Pavlov, Gabriel Goh, Scott Gray, Chelsea Voss, Alec Rad- ford, Mark Chen, and Ilya Sutskever. 2021. Zero-Shot Text-to-Image Generation. In International Conference on Machine Learning . PMLR, 8821–8831

    2021

  31. [39]

    Abhik Roychoudhury. 2024. Program Repair and Trusted Automatic Program- ming (ISEC ’24). Association for Computing Machinery, New York, NY, USA, Article 2, 1 pages

    2024

  32. [40]

    Baptiste Roziere, Jonas Gehring, Fabian Gloeckle, Sten Sootla, Itai Gat, Xi- aoqing Ellen Tan, Yossi Adi, Jingyu Liu, Romain Sauvestre, Tal Remez, et al

  33. [41]

    Ben Shneiderman. 2007. Creativity support tools: accelerating discovery and innovation. Commun. ACM 50, 12 (2007), 20–32

    2007

  34. [42]

    Dudley Storey. 2012. Tools, Technologies, and the Future of CSS Animation. Apress, Berkeley, CA, 155–165

    2012

  35. [43]

    Jakob Tholander and Martin Jonsson. 2023. Design Ideation with AI - Sketching, Thinking and Talking with Generative Machine Learning Models. In Proceedings of the 2023 ACM Designing Interactive Systems Conference (Pittsburgh, PA, USA) (DIS ’23). Association for Computing Machi...

    2023

  36. [44]

    Glassman, Jeevana Priya Inala, and Chenglong Wang

    Priyan Vaithilingam, Elena L. Glassman, Jeevana Priya Inala, and Chenglong Wang. 2024. DynaVis: Dynamically Synthesized UI Widgets for Visualization Editing. In Proceedings of the CHI Conference on Human Factors in Computing Systems (CHI ’24). Association for Computing Machine...

    2024

  37. [45]

    Lei Wang, Songheng Zhang, Yun Wang, Ee-Peng Lim, and Yong Wang. 2023. LLM4Vis: Explainable Visualization Recommendation Using ChatGPT. arXiv preprint (2023). arXiv:2310.07652

    2023 arXiv

  38. [46]

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. 2023. Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. arXiv:2201.11903 [cs.CL]

    2023 arXiv

  39. [47]

    Ziming Wu, Yulun Jiang, Yiding Liu, and Xiaojuan Ma. 2020. Predicting and diagnosing user engagement with mobile ui animation via a data-driven approach. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (CHI ’20). Association for Computing Machin...

    2020

  40. [48]

    Zhijie Xia, Kyzyl Monteiro, Kevin Van, and Ryo Suzuki. 2023. RealityCanvas: augmented reality sketching for embedded and responsive scribble animation effects. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology (UIST ’23) . Association fo...

    2023

  41. [49]

    Ping Zhang. 1999. Will You Use Animation on Your Web Pages? Experiments of Animation Effects and Implications for Web User Interface Design and Online Advertising. In Doing Business on the Internet: Opportunities and Pitfalls . Springer, 35–51

    1999

  42. [50]

    Jiayi Zhou, Renzhong Li, Junxiu Tang, Tan Tang, Haotian Li, Weiwei Cui, and Yingcai Wu. 2024. Understanding Nonlinear Collaboration Between Human and AI Agents: A Co-design Framework for Creative Design. In Proceedings of the CHI Conference on Human Factors in Computing System...

    2024

  43. [2023]

    arXiv preprint (2023)

    Code Llama: Open Foundation Models for Code. arXiv preprint (2023). AnyAni: An Interactive System with Generative AI for Animation Effect Creation and Code Understanding in Web DevelopmentConference’17, July 2017, Washington, DC, USA arXiv:2308.12950

    2023 arXiv

  44. [2024]

    arXiv:2408.16500 [cs.CV]

    CogVLM2: Visual Language Models for Image and Video Understanding. arXiv:2408.16500 [cs.CV]

    arXiv

Pith tools

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