Purrfessor: A Fine-tuned Multimodal LLaVA Diet Health Chatbot

Chuan Tian; Dhavan Shah; Linqi Lu; Sijia Yang; Yifan Deng

REVIEW 3 major objections 5 minor 27 references

Purrfessor: A Fine-tuned Multimodal LLaVA Diet Health Chatbot

T0 review · 3 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Purrfessor, a fine-tuned LLaVA chatbot with a cat persona, increases perceived care and interest in dietary advice compared to a GPT-4 bot, but not compliance intentions.

desk verdict A concrete LLaVA fine-tuning pipeline and a small user pilot with a reasonable question, but the headline causal claims are under-identified and the performance evaluation appears to leak training data. read the letter →

arxiv 2411.14925 v1 pith:5LNODWGK submitted 2024-11-22 cs.HC cs.AI

Linqi Lu , Yifan Deng , Chuan Tian , Sijia Yang , Dhavan Shah This is my paper · ORCID

classification cs.HCcs.AI

keywords dietarychatbotmultimodalLLaVALow-RankAdaptationanthropomorphicpersonauserengagementguidancehuman-in-the-loop

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 introduces Purrfessor, a dietary-guidance chatbot built by fine-tuning the multimodal LLaVA model on food and nutrition data, giving it a cat-like persona and an interface that accepts images of meals. The central claim, tested in a 2 by 3 between-subjects experiment, is that this anthropomorphic chatbot makes users feel more cared for (β = 1.59, p = 0.04) and more interested (β = 2.26, p = 0.01) than a GPT-4 bot, while not significantly changing their intention to follow its advice. The paper also reports a simulation study in which the fine-tuned model detects foods accurately and answers clearly, though with terser responses than GPT-4. Why it matters: health chatbots often fail to hold users' attention, so if a persona and domain fine-tuning can raise engagement, that points a way toward more effective digital dietary coaching.

What carries the argument

The machinery is a pipeline: LLaVA (a multimodal model pairing CLIP's visual encoder with a Vicuna language decoder) is fine-tuned with Low-Rank Adaptation (LoRA) on an instruction dataset built from USDA FoodData Central, Recipe1M, and 500 human-annotated food images, so the model can answer questions about uploaded meal photos. The other moving part is the pet profile: a cat avatar with glasses and a bowtie, presented in a web UI with prompt suggestions, image upload, tooltips, and an onboarding walkthrough. The experiment varies profile (bot vs pet) and model (GPT-4 vs raw LLaVA vs fine-tuned LLaVA) to isolate the contribution of each, measuring care, interest, user-experience quality, satisfaction, and compliance intention.

What would settle it

Run the same 2x3 experiment with response length and formatting controlled across conditions—for example, by rewriting all outputs to the same verbosity and adding the same greetings and closings—and check whether Purrfessor's advantages in care and interest survive; if they disappear, the effect is driven by response style, not by the persona or the fine-tuned model.

Watch

Extended reading notes

Core claim

The paper's central discovery is that a cat-persona, fine-tuned LLaVA chatbot can shift users' subjective experience of a health chatbot: relative to a GPT-4 bot, Purrfessor significantly raised perceived care and user interest in the regression analyses, and a raw LLaVA cat chatbot did so as well for both outcomes. These engagement gains did not carry over to compliance intentions, which showed no significant effect. The paper attributes the gains to the combination of an anthropomorphic visual profile and the multimodal, domain-tuned model, and it complements the experiment with qualitative interviews calling for better responsiveness, personalization, and onboarding guidance.

Load-bearing premise

The causal claim that Purrfessor's design caused the higher care and interest scores assumes the three model conditions differed only in model and profile, but they also produced different response content (GPT-4 verbose, LLaVA terse), so the measured effects could come from response style rather than from the cat avatar or the fine-tuning.

Editorial extensions

If this is right

Adding an anthropomorphic persona to a domain-tuned chatbot can raise perceived warmth and engagement without changing the underlying advice.
Because fine-tuning used LoRA on a 13B model, similar in-house health chatbots can be built with modest GPU resources rather than relying solely on large API vendors.
The null compliance result implies that engagement alone does not push users to act; future versions need behavior-change hooks such as reminders, planning, or follow-ups.
The concise LLaVA style may be an asset for low-literacy or time-pressed users, though it sacrifices the richer context GPT-4 provides.

Reading between the lines

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

The care and interest effects may partly reflect response style: GPT-4 gave elaborate contextual replies while LLaVA gave short factual lists, so a replication that matches response length and formatting across conditions is needed to separate persona effects from style effects.
The cat persona's charm may be a novelty effect; a longitudinal study would show whether engagement persists or fades after several sessions.
The fine-tuning recipe—public food data plus human-annotated images plus LoRA—transfers to other advice domains, such as physical activity or medication adherence, where image input and concise guidance matter.
Because the sample was 51 undergraduates at one university, the effect sizes may not generalize; a broader sample could test whether the persona works across age and socioeconomic groups, which the paper itself flags.

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.

Desk Editor's Note

read the letter

Colleague, quick take on arXiv:2411.14925. The paper is a concrete system description plus a small pilot experiment. The fine-tuning pipeline (LoRA on LLaVA-13b, USDA FoodData Central, Recipe1M, 500 human-annotated images) is clearly presented, and the 2x3 profile-by-model design is a plausible way to ask whether a cat avatar and fine-tuning move user experience. The qualitative interviews are useful. But the evidence for the headline claims is fragile.

The biggest problem is that Study A's simulation evaluation appears to run on the same 500 human-annotated images used for training (Section 3.1 vs Section 4.1.1). That makes the automated overlap score and the human validation scores uninterpretable as generalization estimates. The "state-of-the-art" and "new benchmark" language in Section 3.3 is not earned by this evaluation.

The user study is also more weakly identified than the abstract suggests. N=51 with 14 predictors, no multiple-comparison correction, no manipulation check results, and an internal p-value mismatch (abstract says p=0.005, body says p=0.01). Care and interest also report identical M and SD, which looks like a copy error. More importantly, the paper only reports dummy contrasts against the GPT-4 bot condition; the 2x3 factorial is never analyzed with profile/model main effects and the interaction. The stress-test note is right: raw LLaVA with cat profile matches or exceeds fine-tuned Purrfessor on care and interest, so the fine-tuning itself contributes nothing detectable in the reported numbers. The response-content confound (GPT-4 gives elaborate contextual answers, LLaVA gives concise factual lists) is documented in the paper itself, so the betas do not isolate the avatar or the fine-tuning.

What is good: the system is real, the pipeline is described in enough detail to be reproduced in principle, and the authors do not oversell the compliance result—they explicitly report F(14,36)=1.25, p=0.29. The interview themes are sensible. But there is no code, data, or model release, which makes external checking hard.

Bottom line: this is a pilot paper with a plausible direction, not a settled result. A serious referee could fix a lot of it by asking for the full factorial analysis, the manipulation check numbers, a corrected abstract, and either a held-out evaluation or a clear statement of the training/eval overlap. I would send it to peer review because the artifact is concrete and the research question is worth pursuing, but it needs major revision before publication. I would not cite it in my own work yet. Recommendation: accept for review, expect heavy revision.

Referee Report

3 major / 5 minor

Summary. The paper introduces Purrfessor, a LLaVA-based multimodal chatbot with a cat avatar, fine-tuned on food and nutrition data, and evaluates it in two studies. Study A reports automated text-overlap scores and human ratings for the fine-tuned model. Study B is a 2 (Profile: Bot vs. Anthropomorphic) × 3 (Model: GPT-4 vs. Raw LLaVA vs. Fine-tuned LLaVA) between-subject experiment (N = 51) with an additional baseline, and the authors claim that the fine-tuned LLaVA cat chatbot (Purrfessor) significantly increased perceived care and interest compared with a GPT-4 bot. Qualitative interviews with eight participants identify responsiveness, personalization, and guidance as key design needs.

Significance. If the causal claims were valid, the paper would provide useful evidence that anthropomorphic profiles and fine-tuned multimodal models can improve user engagement in dietary chatbots. The paper has tangible strengths: it presents a working system, includes human-in-the-loop annotation, uses a randomized factorial design, and reports inter-coder reliability for Study A. However, the significance is currently not established because Study A likely evaluates the model on the same images used for training, and Study B's analysis does not isolate the effects of the cat profile, the base model, the fine-tuning, or the response style. The paper is best viewed as a preliminary system description plus a pilot user study, not as a definitive causal demonstration.

major comments (3)

[Section 4.1.1 vs. Section 3.1] The Study A evaluation set appears to overlap with the training set. Section 3.1 lists a 'Human-Annotated Dataset: A dataset of 500 human-annotated food-related images' used for training, and Section 4.1.1 states that the Study A dataset 'consisted of 500 real-world images sourced via the Google Image Search API' with no explicit statement that these are disjoint sets. If the same 500 images are used for both training and evaluation, the reported text-overlap score (0.67) and human validation scores (e.g., Correctness M = 7.87) reflect memorization rather than generalization. The manuscript must specify the relationship between the two sets or re-run Study A on a held-out set before the performance claims can be accepted.
[Sections 5.1 and 5.4] The causal attribution to the cat persona or fine-tuning is under-identified. The 2 × 3 between-subject design is analyzed only through dummy-variable contrasts against the GPT-4 bot reference; no main effects of Profile or Model, and no Profile × Model interaction, are reported. The Purrfessor condition differs from the reference simultaneously in profile (cat), base model (LLaVA vs. GPT-4), fine-tuning, and response content, and Section 4.2.2 documents that GPT-4 gives elaborate contextual responses while LLaVA gives concise factual lists. The reported estimates themselves undermine the fine-tuning attribution: for care, raw LLaVA cat (β = 1.58, p = 0.02) is nearly identical to fine-tuned LLaVA cat (β = 1.59, p = 0.04), and for interest, raw LLaVA cat has a larger coefficient (β = 2.50 vs. β = 2.26). Please report the factorial contrasts and either provide a content-controlled comparison or substantially weaken the causal claims.
[Sections 5.3 and 5.4, and Abstract] The manipulation check described in Section 5.3 is never reported; we do not know whether participants correctly recalled the chatbot name and profile image, which is a prerequisite for interpreting the profile manipulation. In addition, with N = 51 and 14 predictors, the multiple regressions and dummy contrasts are vulnerable to inflated Type I error; no multiple-comparison correction or confidence intervals are provided. There is also an inconsistency in the reported p-value for interest: the abstract states p = 0.01 (in the first abstract) and p = 0.005 (in the full-text abstract), while Section 5.4 reports p = 0.01. Please clarify the exact p-values and report corrected or adjusted results.

minor comments (5)

[Section 5.3] The Care Perceptions and User Interest scales are reported with identical descriptive statistics (M = 4.97, SD = 1.33, α = .91), which is likely a copy-and-paste error; please verify the actual values.
[Section 5.1] The study design mentions an 'additional Baseline condition (ChatGPT only)', but the analysis in Section 5.4 does not clarify how this baseline relates to the 'GPT-4 bot' condition or whether it was included in the regressions.
[Section 4.1.2] The text-overlap metric and its 0.6 threshold are described only loosely; please provide the exact formula for the overlap score and a justification for the threshold.
[Section 5.5] The qualitative interview section would benefit from more detail about participant selection, interview length, the coding procedure, and how thematic saturation was assessed.
[Section 2.2.4] The statement that ConversationDB can 'support model fine-tuning by storing real user interaction data' is not connected to any reported use in the studies; please either clarify or remove this claim.

Circularity Check

1 steps flagged · score 6.0 of 10

Study A evaluates the fine-tuned model on the same 500 Google Image Search images used for training, making the performance numbers partly in-sample; Study B's user-experiment results are independent, so circularity is partial.

fitted input called prediction [Section 3.1 (Training Dataset) and Section 4.1.1 (Study A Dataset)]
"Section 3.1: 'Human-Annotated Dataset: A dataset of 500 human-annotated food-related images was included to enhance the model’s capability to provide healthy cooking instructions starting from visual inputs of ingredients.' Section 4.1.1: 'The dataset for this study consisted of 500 real-world images sourced via the Google Image Search API, paired with synthetically generated prompts designed to simulate a wide range of user queries.'"

The only 500-image Google Image Search dataset described in the paper is the human-annotated training set from Section 3.1/3.2, and Section 4.1.1 evaluates 'the dataset for this study' as 500 Google Image Search images without introducing any separate held-out collection. Because the model was instruction-tuned on those images' GPT-4o-generated captions and Q&As (Section 3.2.2) and human-reviewed answers, the automated overlap scores (4.2.1) and human validation ratings (4.2.3) measure in-sample reconstruction of the training labels rather than out-of-sample prediction. The reported performance numbers are therefore forced in part by construction: the model is being credited for reproducing inputs it was fitted to.

full rationale

The strongest circularity is confined to Study A. The paper's training dataset includes 500 human-annotated Google Image Search images, and Study A's evaluation dataset is also described as 500 real-world images sourced via the Google Image Search API, with no second collection described; thus the model-performance evaluation appears to run on the training set. That makes the overlap scores and human validation partly in-sample. Study B, by contrast, is an independent between-subjects user experiment with new participants and new dependent measures, so the central user-experience claim about care and interest does not reduce to a fitted input. The user-experiment analysis is under-identified (the 2x3 factorial is never analyzed with main effects or interactions, and raw LLaVA-cat shows similar or larger betas than the fine-tuned cat), but that is a causal-attribution/correctness problem rather than a circularity problem under the stated rules. No load-bearing self-citation or imported uniqueness theorem is present. Overall: partial circularity, score 6.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

All assumptions are domain assumptions from the system and study design; no mathematical axioms are introduced. The only hand-chosen numeric threshold is the overlap cutoff. The cat persona is an interface element, not a postulated theoretical entity, so it is not listed as an invented entity.

free parameters (1)

text overlap threshold = 0.6
Chosen without calibration in Section 4.1.2 to flag low-performing cases; it determines which 100 cases receive human review and therefore shapes the qualitative conclusions.

assumptions (4)

domain assumption LLaVA-v1.6-13b can learn food and nutrition behavior from the constructed instruction dataset.
Section 3 treats the base model plus LoRA fine-tuning as sufficient; no ablation against other backbones is reported.
domain assumption GPT-4o-generated captions and Q&A pairs are accurate enough to serve as training targets.
Section 3.2.2 uses GPT-4o outputs as ground truth after human review; errors in these targets would propagate into the fine-tuned model.
domain assumption The self-report scales validly measure care, interest, satisfaction, and compliance intention.
Section 5.3 relies on Likert-scale questionnaires with Cronbach's alpha; no validation against behavior or external criteria is provided.
domain assumption A 15-minute single conversation is enough to measure stable user perceptions.
Section 5.2 assigns 15-minute interactions; whether this captures lasting attitudes is not established.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Purrfessor: A Fine-tuned Multimodal LLaVA Diet Health Chatbot." pith.science (2026). https://pith.science/paper/5LNODWGK

@misc{pith2026241114925,
  author       = {Pith},
  title        = {Pith review of: Purrfessor: A Fine-tuned Multimodal LLaVA Diet Health Chatbot},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5LNODWGK}},
  note         = {Machine review of arXiv:2411.14925}
}

abstract

This study introduces Purrfessor, an innovative AI chatbot designed to provide personalized dietary guidance through interactive, multimodal engagement. Leveraging the Large Language-and-Vision Assistant (LLaVA) model fine-tuned with food and nutrition data and a human-in-the-loop approach, Purrfessor integrates visual meal analysis with contextual advice to enhance user experience and engagement. We conducted two studies to evaluate the chatbot's performance and user experience: (a) simulation assessments and human validation were conducted to examine the performance of the fine-tuned model; (b) a 2 (Profile: Bot vs. Pet) by 3 (Model: GPT-4 vs. LLaVA vs. Fine-tuned LLaVA) experiment revealed that Purrfessor significantly enhanced users' perceptions of care ($\beta = 1.59$, $p = 0.04$) and interest ($\beta = 2.26$, $p = 0.01$) compared to the GPT-4 bot. Additionally, user interviews highlighted the importance of interaction design details, emphasizing the need for responsiveness, personalization, and guidance to improve user engagement.

Figures

Figures reproduced from arXiv: 2411.14925 by the authors.

**Figure 1.** System Architecture. 2.3. Chatbot Profile The Purrfessor chatbot’s visual persona is represented by a cat-themed avatar with large, inviting eyes, glasses, and a bowtie, projecting a scholarly yet charming personality. This profile is designed to create a warm and engaging experience, encouraging users to return for guidance on meal planning. 2.4. Interface Design and Interaction Flow 2.4.1 Conversation Bar with Pr… view at source ↗

**Figure 2.** Purrfessor Nov 24 Version User Interface Vicuna language decoder” to bridge visual and linguistic understanding in multimodal contexts [14]. This architecture allows LLaVA to process and interpret both images and text, enabling it to handle diverse instruction types—such as image descriptions, dialogue, and complex reasoning—by leveraging the strengths of both its visual and language model components. During our … view at source ↗

**Figure 3.** The Multiple Linear Regression Results 5.5.1 Interaction and Responsiveness Participants expressed a desire for more seamless and dynamic interactions. Long response times disrupted engagement, reducing the fluidity of conversations. Suggestions included displaying user input immediately and generating chatbot responses progressively to enhance real-time interaction. “The waiting time is long. You can make it sam… view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Purrfessor April 24 Version User Interface. 6. Discussion and Conclusion Our simulation test reveals a distinction in descriptive nuance between GPT-4 and LLaVA. While GPT-4 generated responses that were rich in context, offering nutritional insights and detailed des…

**Figure 5.** Figure 5: Chatbot Purrfessor Q&A Conversation Demo. [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 26 canonical work pages

[1]

Beck and J

D. Beck and J. Lin. Efficient web scraping with python and regular expressions. Journal of Data Science , 18(3):234– 245, 2020. 4

work page 2020
[2]

T. W. Bickmore, L. Caruso, K. Clough-Gorr, and T. Heeren. ”it’s just like you talk to a friend”: Relational agents for older adults. Interacting with Computers, 17(6):711–735, 2005. 1, 2

work page 2005
[3]

Using thematic analysis in psychology

Virginia Braun and Victoria Clarke. Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2):77– 101, 2006. 7

work page 2006
[4]

Thematic analysis: A practical guide

Virginia Braun and Victoria Clarke. Thematic analysis: A practical guide. SAGE Publications Ltd., 2019. 8

work page 2019
[5]

Cecchini, F

M. Cecchini, F. Sassi, J. A. Lauer, Y . Y . Lee, V . Guajardo- Barron, and D. Chisholm. Tackling unhealthy diets, physical inactivity, and obesity: Health effects and cost-effectiveness. The Lancet, 376(9754):1775–1784, 2010. 1

work page 2010
[6]

Nearly half of americans use dig- ital voice assistants, mostly on their smartphones, 2017

Pew Research Center. Nearly half of americans use dig- ital voice assistants, mostly on their smartphones, 2017. Retrieved from https://www.pewresearch.org/ fact - tank / 2017 / 12 / 12 / nearly - half - of - americans- use- digital- voice- assistants- mostly-on-their-smartphones/ . 1

work page 2017
[7]

Fadhil and S

A. Fadhil and S. Gabrielli. Addressing challenges in pro- moting healthy lifestyles: The al-chatbot approach. In Pro- ceedings of the 11th EAI International Conference on Per- vasive Computing Technologies for Healthcare, pages 261– 265, 2017. 2

work page 2017
[8]

D. D. Farhud. Impact of lifestyle on health. Iranian Journal of Public Health, 44(11):1442, 2015. 1

work page 2015

Show all 27 references

[9]

Google custom search json api, 2023

Google. Google custom search json api, 2023. Re- trieved from https://developers.google.com/ custom-search/v1/overview. 4

2023
[10]

R. Han, A. Todd, S. Wardak, S. R. Partridge, and R. Raeside. Feasibility and acceptability of chatbots for nutrition and physical activity health promotion among adolescents: Sys- tematic scoping review with adolescent consultation. JMIR Human Factors, 10(1):e43227, 2023. 2, 8

2023
[11]

J.-W. Hong. I was born to love ai: The influence of social sta- tus on ai self-efficacy and intentions to use ai. International Journal of Communication, 16:20, 2022. 7

2022
[12]

L. C. Klopfenstein, S. Delpriori, S. Malatini, and A. Bogli- olo. The rise of bots: A survey of conversational interfaces, patterns, and paradigms. In Proceedings of the 2017 Con- ference on Designing Interactive Systems , pages 555–565,

2017
[13]

Laranjo, A

L. Laranjo, A. G. Dunn, H. L. Tong, A. B. Kocaballi, J. Chen, R. Bashir, and E. Coiera. Conversational agents in health- care: A systematic review. Journal of the American Medical Informatics Association, 25(9):1248–1258, 2018. 1

2018
[14]

H. Liu, C. Li, Q. Wu, and Y . J. Lee. Visual instruction tun- ing. In Advances in Neural Information Processing Systems, pages 34892–34916, 2024. 1, 3

2024
[15]

C. A. Maher, C. R. Davis, R. G. Curtis, C. E. Short, and K. J. Murphy. A physical activity and diet program deliv- ered by an artificially intelligent virtual health coach: Proof- of-concept study. JMIR mHealth and uHealth, 8(7):e17558,
[16]

A. S. Miner, L. Laranjo, and A. B. Kocaballi. Chatbots in the fight against the covid-19 pandemic. npj Digital Medicine, 3 (1):1–4, 2020. 2

2020
[17]

Nazary, Y

F. Nazary, Y . Deldjoo, and T. Di Noia. Chatgpt- healthprompt: Harnessing the power of xai in prompt-based healthcare decision support using chatgpt. InEuropean Con- ference on Artificial Intelligence , pages 382–397. Springer,
[18]

Pereira and ´O

J. Pereira and ´O. D´ıaz. Using health chatbots for behavior change: A mapping study. Journal of Medical Systems , 43 (5):135, 2019. 2, 5, 8

2019
[19]

Radford, J

A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, and I. Sutskever. Learning transferable visual models from nat- ural language supervision. In Proceedings of the Inter- national Conference on Machine Learning , pages 14108– 14119, 2021. 4

2021
[20]

Redmon and A

J. Redmon and A. Farhadi. Yolov3: An incremental improve- ment, 2018. arXiv preprint arXiv:1804.02767. 4

2018 arXiv
[21]

Salvador, N

A. Salvador, N. Hynes, Y . Aytar, J. Marin, F. Ofli, I. We- ber, and A. Torralba. Learning cross-modal embeddings for cooking recipes and food images. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recog- nition, pages 3020–3028, 2017. 1, 3

2017
[22]

On the importance of ux quality as- pects for different product categories

Martin Schrepp, Jessica Kollmorgen, Anna-Lena Meiners, Andreas Hinderks, Dominique Winter, Harry B Santoso, and J¨org Thomaschewski. On the importance of ux quality as- pects for different product categories. International Journal of Interactive Multimedia and Artificial Intel...

2023
[23]

Shyam Sundar and Y

S. Shyam Sundar and Y . A. Marathe. Personalization ver- sus customization: The importance of agency, privacy, and power usage. Human Communication Research, 36(3):298– 322, 2010. 8

2010
[24]

Wang and Y .-S

Y .-Y . Wang and Y .-S. Wang. Development and validation of an artificial intelligence anxiety scale: An initial application in predicting motivated learning behavior.Interactive Learn- ing Environments, 30(4):619–634, 2022. 7

2022
[25]

Wiltshire and Selina Ronkainen

William A. Wiltshire and Selina Ronkainen. Enhancing user experience in conversational agents through personal- ized interaction and adaptive design. International Journal of Human-Computer Studies, 147:102567, 2021. 7, 8

2021
[26]

Zhang, Y

J. Zhang, Y . J. Oh, P. Lange, Z. Yu, and Y . Fukuoka. Artifi- cial intelligence chatbot behavior change model for design- ing ai chatbots to promote physical activity and a healthy diet. Journal of Medical Internet Research , 22(9):e22845,
[27]

L. Zhu, W. Wu, and Y . Wang. Techniques for image search and retrieval in social media studies. International Journal of Media and Data Science, 6(2):112–130, 2021. 4 Purrfessor: A Fine-tuned LLaV A Diet Health Chatbot Supplementary Material Appendix A: Validation Criteria Tab...

2021

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Beck and J

D. Beck and J. Lin. Efficient web scraping with python and regular expressions. Journal of Data Science , 18(3):234– 245, 2020. 4

work page 2020

[2] [2]

T. W. Bickmore, L. Caruso, K. Clough-Gorr, and T. Heeren. ”it’s just like you talk to a friend”: Relational agents for older adults. Interacting with Computers, 17(6):711–735, 2005. 1, 2

work page 2005

[3] [3]

Using thematic analysis in psychology

Virginia Braun and Victoria Clarke. Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2):77– 101, 2006. 7

work page 2006

[4] [4]

Thematic analysis: A practical guide

Virginia Braun and Victoria Clarke. Thematic analysis: A practical guide. SAGE Publications Ltd., 2019. 8

work page 2019

[5] [5]

Cecchini, F

M. Cecchini, F. Sassi, J. A. Lauer, Y . Y . Lee, V . Guajardo- Barron, and D. Chisholm. Tackling unhealthy diets, physical inactivity, and obesity: Health effects and cost-effectiveness. The Lancet, 376(9754):1775–1784, 2010. 1

work page 2010

[6] [6]

Nearly half of americans use dig- ital voice assistants, mostly on their smartphones, 2017

Pew Research Center. Nearly half of americans use dig- ital voice assistants, mostly on their smartphones, 2017. Retrieved from https://www.pewresearch.org/ fact - tank / 2017 / 12 / 12 / nearly - half - of - americans- use- digital- voice- assistants- mostly-on-their-smartphones/ . 1

work page 2017

[7] [7]

Fadhil and S

A. Fadhil and S. Gabrielli. Addressing challenges in pro- moting healthy lifestyles: The al-chatbot approach. In Pro- ceedings of the 11th EAI International Conference on Per- vasive Computing Technologies for Healthcare, pages 261– 265, 2017. 2

work page 2017

[8] [8]

D. D. Farhud. Impact of lifestyle on health. Iranian Journal of Public Health, 44(11):1442, 2015. 1

work page 2015