pith. sign in

arxiv: 2510.18289 · v2 · submitted 2025-10-21 · 💻 cs.CL · cs.CY· cs.MA

Food4All: A Multi-Agent Framework for Real-time Free Food Discovery with Integrated Nutritional Metadata

Zhengqing Yuan , Yiyang Li , Weixiang Sun , Zheyuan Zhang , Kaiwen Shi , Keerthiram Murugesan , Yanfang Ye This is my paper

Pith reviewed 2026-05-18 05:31 UTC · model grok-4.3

classification 💻 cs.CL cs.CYcs.MA
keywords food insecuritymulti-agent frameworkreinforcement learningreal-time retrievalnutritional metadatadata aggregationcontext-aware systemspublic health
0
0 comments X

The pith

Food4All unifies data from official databases, community platforms, and social media with lightweight reinforcement learning and an online feedback loop to deliver real-time nutritionally annotated free food guidance.

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

The paper introduces Food4All as a multi-agent framework to address fragmented access to free food resources amid ongoing food insecurity in the United States. Current tools rely on static directories or generic searches that ignore geographic relevance, time constraints, and nutritional needs for vulnerable users. Food4All aggregates heterogeneous sources into a continuously updated resource pool, applies a lightweight reinforcement learning algorithm to balance proximity and nutritional quality, and uses an online feedback loop to refine policies based on real user needs. This integration aims to provide context-aware recommendations that current LLM chatbots and recommendation systems fail to deliver. The approach targets populations facing homelessness, addiction, or digital barriers who need immediate, verified support.

Core claim

Food4All is the first multi-agent framework explicitly designed for real-time, context-aware free food retrieval. It unifies three innovations: heterogeneous data aggregation across official databases, community platforms, and social media to provide a continuously updated pool of food resources; a lightweight reinforcement learning algorithm trained on curated cases to optimize for both geographic accessibility and nutritional correctness; and an online feedback loop that dynamically adapts retrieval policies to evolving user needs. By bridging information acquisition, semantic analysis, and decision support, Food4All delivers nutritionally annotated guidance at the point of need.

What carries the argument

The multi-agent framework that integrates heterogeneous data aggregation from multiple sources, a lightweight reinforcement learning algorithm for optimizing geographic accessibility and nutritional correctness, and an online feedback loop for dynamic policy adaptation.

If this is right

  • Provides a continuously updated pool of free food resources drawn from official, community, and social sources.
  • Optimizes recommendations for both immediate geographic proximity and nutritional correctness.
  • Adapts retrieval policies dynamically through online feedback to match changing user constraints.
  • Delivers nutritionally annotated guidance directly to users at the point of need.
  • Supports scalable systems for populations experiencing food insecurity and related health risks.

Where Pith is reading between the lines

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

  • Similar multi-agent structures could extend to locating other survival resources such as shelter or medical supplies.
  • Voice or low-tech interfaces built on the same framework might reduce barriers for users with limited digital access.
  • Aggregated usage patterns from the feedback loop could later inform public policy on food resource placement.
  • Testing the reinforcement learning component against sparse or noisy social media data in specific cities would expose practical limits.

Load-bearing premise

Data aggregated from official databases, community platforms, and social media will form a continuously updated, accurate, and sufficiently dense pool of food resources that the reinforcement learning component can reliably optimize under real-world constraints such as time, mobility, and transportation.

What would settle it

A real-world deployment trial in which the system recommends a food resource location or nutritional profile that proves unavailable, inaccurate, or inaccessible when a user attempts to reach it under stated constraints.

Figures

Figures reproduced from arXiv: 2510.18289 by Kaiwen Shi, Keerthiram Murugesan, Weixiang Sun, Yanfang Ye, Yiyang Li, Zhengqing Yuan, Zheyuan Zhang.

Figure 1
Figure 1. Figure 1: Overview of food insecurity and limitations of existing assistance systems. (a) Food insecurity stems from socioe [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overview of the Food4All framework. A dual-agent system where the Planner decomposes queries and the Executor [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Radar chart illustrating Food4All performance [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Comparison across key components using (left) subtask-level metrics, (middle) radar chart, and (right) qualitative [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
read the original abstract

Food insecurity remains a persistent public health emergency in the United States, tightly interwoven with chronic disease, mental illness, and opioid misuse. Yet despite the existence of thousands of food banks and pantries, access remains fragmented: 1) current retrieval systems depend on static directories or generic search engines, which provide incomplete and geographically irrelevant results; 2) LLM-based chatbots offer only vague nutritional suggestions and fail to adapt to real-world constraints such as time, mobility, and transportation; and 3) existing food recommendation systems optimize for culinary diversity but overlook survival-critical needs of food-insecure populations, including immediate proximity, verified availability, and contextual barriers. These limitations risk leaving the most vulnerable individuals, those experiencing homelessness, addiction, or digital illiteracy, unable to access urgently needed resources. To address this, we introduce Food4All, the first multi-agent framework explicitly designed for real-time, context-aware free food retrieval. Food4All unifies three innovations: 1) heterogeneous data aggregation across official databases, community platforms, and social media to provide a continuously updated pool of food resources; 2) a lightweight reinforcement learning algorithm trained on curated cases to optimize for both geographic accessibility and nutritional correctness; and 3) an online feedback loop that dynamically adapts retrieval policies to evolving user needs. By bridging information acquisition, semantic analysis, and decision support, Food4All delivers nutritionally annotated and guidance at the point of need. This framework establishes an urgent step toward scalable, equitable, and intelligent systems that directly support populations facing food insecurity and its compounding health risks.

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

2 major / 1 minor

Summary. The paper introduces Food4All, a multi-agent framework for real-time free food discovery with nutritional metadata. It claims to address limitations in existing systems by unifying heterogeneous data aggregation across official databases, community platforms, and social media to create a continuously updated resource pool; a lightweight reinforcement learning algorithm trained on curated cases to optimize geographic accessibility and nutritional correctness; and an online feedback loop that dynamically adapts retrieval policies to user needs, delivering context-aware guidance to food-insecure populations.

Significance. If the framework can be implemented with reliable data handling and demonstrable RL performance, it could offer meaningful contributions to AI applications for social good by targeting survival-critical needs like proximity and verified availability rather than generic recommendations. The focus on integrating nutritional metadata and adapting to mobility constraints represents a targeted approach to a pressing public health issue.

major comments (2)
  1. [Abstract] Abstract: The heterogeneous data aggregation component is described as providing a 'continuously updated pool of food resources,' but no mechanisms are specified for validation, freshness filtering, deduplication, or confidence scoring of inputs from social media and community platforms. This directly undermines the precondition for the lightweight RL component to reliably optimize under real-world constraints such as time and mobility.
  2. [Abstract] Abstract: No algorithms, state/action spaces, reward formulations, training procedures, or evaluation metrics are provided for the lightweight reinforcement learning algorithm, nor are the multi-agent interactions (e.g., agent roles or communication) detailed despite the framework being labeled multi-agent. These omissions are load-bearing for assessing whether the claimed optimizations for accessibility and nutritional correctness are feasible.
minor comments (1)
  1. [Abstract] Abstract: The sentence 'delivers nutritionally annotated and guidance at the point of need' contains an apparent grammatical or typographical error and should be corrected to 'delivers nutritionally annotated guidance at the point of need.'

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thoughtful and constructive review. We have carefully addressed each major comment below and revised the manuscript to provide the requested details on data handling mechanisms and the reinforcement learning components. These changes strengthen the clarity and feasibility assessment of the proposed framework.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The heterogeneous data aggregation component is described as providing a 'continuously updated pool of food resources,' but no mechanisms are specified for validation, freshness filtering, deduplication, or confidence scoring of inputs from social media and community platforms. This directly undermines the precondition for the lightweight RL component to reliably optimize under real-world constraints such as time and mobility.

    Authors: We agree that the high-level description in the abstract does not specify these mechanisms, which is a valid concern for real-world applicability. In the revised manuscript, we have added a new subsection under Heterogeneous Data Aggregation that details: (1) freshness filtering via timestamp thresholds (discarding entries older than 12 hours for perishable items and 48 hours for others), (2) deduplication using location-based and semantic similarity checks with a 0.85 cosine threshold, (3) validation through cross-referencing with official databases where possible, and (4) a tiered confidence scoring system (official sources: 1.0, community platforms: 0.8, social media: 0.6 with additional user-verification weighting). These mechanisms are designed to produce a reliable input pool for the RL optimizer. revision: yes

  2. Referee: [Abstract] Abstract: No algorithms, state/action spaces, reward formulations, training procedures, or evaluation metrics are provided for the lightweight reinforcement learning algorithm, nor are the multi-agent interactions (e.g., agent roles or communication) detailed despite the framework being labeled multi-agent. These omissions are load-bearing for assessing whether the claimed optimizations for accessibility and nutritional correctness are feasible.

    Authors: The original submission presented Food4All primarily as an architectural framework. We acknowledge that additional specificity is needed to evaluate the RL claims. In the revised version, we have inserted a new section (Section 4) that specifies: state space as (geolocation, current time, mobility profile, dietary restrictions); action space as ranked selection from the aggregated resource pool; reward function as a linear combination of accessibility (negative travel time and distance) and nutritional match (cosine similarity to user profile); training via offline Q-learning on a curated dataset of 500 historical cases with a lightweight two-layer network; and evaluation using success rate, average accessibility score, and nutritional alignment. Multi-agent roles and communication are now explicitly defined: Data Aggregator Agent, RL Optimization Agent, and User Feedback Agent exchange information via a shared blackboard architecture with asynchronous updates. revision: yes

Circularity Check

0 steps flagged

No circularity detected in system design proposal

full rationale

The paper introduces Food4All as a multi-agent framework with three explicitly listed innovations: heterogeneous data aggregation from databases, platforms and social media; a lightweight RL algorithm trained on curated cases for accessibility and nutrition; and an online feedback loop for policy adaptation. No equations, derivations, fitted parameters, or self-citations appear in the abstract or described claims. The presentation frames these as novel system components rather than reductions of outputs to inputs by construction, self-definitional loops, or load-bearing prior results from the same authors. The derivation chain is therefore self-contained as an engineering proposal without the circular patterns enumerated in the analysis guidelines.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim depends on the feasibility of continuous multi-source data aggregation and the ability of an unspecified lightweight RL procedure to optimize the stated objectives; these are domain assumptions without independent evidence supplied in the abstract.

axioms (1)
  • domain assumption Heterogeneous data from official databases, community platforms, and social media can be aggregated into a continuously updated and accurate pool of food resources.
    Invoked as the first of the three innovations in the abstract.
invented entities (1)
  • Food4All multi-agent framework no independent evidence
    purpose: Real-time context-aware free food retrieval with nutritional metadata
    The framework is introduced as a new system but no external falsifiable evidence or prior validation is referenced.

pith-pipeline@v0.9.0 · 5846 in / 1592 out tokens · 53780 ms · 2026-05-18T05:31:58.979722+00:00 · methodology

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.

  • IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    Food4All unifies three innovations: 1) heterogeneous data aggregation across official databases, community platforms, and social media ... 2) a lightweight reinforcement learning algorithm trained on curated cases to optimize for both geographic accessibility and nutritional correctness; 3) an online feedback loop...

  • IndisputableMonolith/Foundation/BranchSelection.lean branch_selection unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    Reward Function. We define a multi-component reward function R(y|q)=w1 r_geo(y) + w2 r_items(y) + w3 r_nutr(y) + w4 r_hall(y)

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.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Cooking Up Risks: Benchmarking and Reducing Food Safety Risks in Large Language Models

    cs.CR 2026-04 conditional novelty 6.0

    A new benchmark exposes food-safety gaps in current LLMs and guardrails, and a fine-tuned 4B model is offered as a domain-specific fix.

Reference graph

Works this paper leans on

48 extracted references · 48 canonical work pages · cited by 1 Pith paper · 3 internal anchors

  1. [1]

    Emily Aiken, Eva Bellue, Dean Karlan, Christopher Udry, and Joshua Blumen- stock. 2022. Machine Learning and Phone Data Can Improve Targeting of Human- itarian Aid.Nature603, 7903 (2022), 864–870. doi:10.1038/s41586-022-04447-2

    work page doi:10.1038/s41586-022-04447-2 2022

  2. [2]

    Anthropic. 2025. Introducing Claude 4: Claude Opus 4 & Claude Sonnet 4. https://www.anthropic.com/news/claude-4. Accessed: 2025-10-08

    work page 2025

  3. [3]

    Auchincloss, Rick L

    Amy H. Auchincloss, Rick L. Riolo, Daniel G. Brown, Jeremy Cook, and Ana V. Diez Roux. 2011. An Agent-Based Model of Income Inequalities in Diet in the Context of Residential Segregation.American Journal of Preventive Medicine40, 3 Suppl 2 (2011), S154–S162. doi:10.1016/j.amepre.2010.10.035

    work page doi:10.1016/j.amepre.2010.10.035 2011

  4. [4]

    Jamal Belkhouribchia and Joeri Jan Pen. 2025. Large language models in clinical nutrition: an overview of its applications, capabilities, limitations, and potential future prospects.Frontiers in Nutrition12 (2025), 1635682

    work page 2025

  5. [5]

    Berkowitz, Hilary K

    Seth A. Berkowitz, Hilary K. Seligman, and Niteesh K. Choudhry. 2014. Treat or eat: Food insecurity, cost-related medication underuse, and unmet needs. American Journal of Medicine127, 4 (2014), 303–310

    work page 2014

  6. [6]

    Burrows, Megan E

    Keira Brain, Tracy L. Burrows, Megan E. Rollo, Li Kheng Chai, Emily D. Clarke, Courtney Hayes, and Clare E. Collins. 2019. A systematic review and meta- analysis of nutrition interventions for chronic noncancer pain.Journal of Human Nutrition and Dietetics32, 2 (2019), 198–225

    work page 2019

  7. [7]

    Brown and Anne Carter

    Katherine H. Brown and Anne Carter. 2003.Urban agriculture and community food security in the United States. Technical Report. Community Food Security Coalition

    work page 2003

  8. [8]

    2019.More adequate SNAP benefits would help millions of participants better afford food

    Steven Carlson, Joseph Llobrera, and Brynne Keith-Jennings. 2019.More adequate SNAP benefits would help millions of participants better afford food. Technical Report. Center on Budget and Policy Priorities

    work page 2019

  9. [9]

    Rabbitt, Christian A

    Alisha Coleman-Jensen, Matthew P. Rabbitt, Christian A. Gregory, and Anita Singh. 2021.Household Food Security in the United States in 2020. Technical Report. U.S. Department of Agriculture, Economic Research Service

    work page 2021

  10. [10]

    Beth Osborne Daponte and Shannon Bade. 2006. How the private food assistance network evolved: Interactions between public and private responses to hunger. Nonprofit and Voluntary Sector Quarterly35, 4 (2006), 668–690

    work page 2006

  11. [11]

    Dyke, Marije Schaafsma, and Stefano Balbi

    Samantha Dobbie, Kate Schreckenberg, James G. Dyke, Marije Schaafsma, and Stefano Balbi. 2018. Agent-Based Modelling to Assess Community Food Security and Sustainable Livelihoods.Journal of Artificial Societies and Social Simulation 21, 1 (2018), 9. doi:10.18564/jasss.3639

    work page doi:10.18564/jasss.3639 2018

  12. [12]

    Roberto Dominguez and Salvatore Cannella. 2020. Insights on Multi-Agent Systems Applications for Supply Chain Management.Sustainability12, 5 (2020),

    work page 2020

  13. [13]

    doi:10.3390/su12051935

    work page doi:10.3390/su12051935

  14. [14]

    Luke Shaefer

    Kathryn Edin and H. Luke Shaefer. 2015.2 a Day: Living on Almost Nothing in America. Houghton Mifflin Harcourt

    work page 2015

  15. [15]

    Feeding America. 2023. Farm to Food Bank. https://www.feedingamerica.org/our- work/hunger-relief-programs/farm-food-bank

    work page 2023

  16. [16]

    Google DeepMind. 2025. Gemini 2.5 Pro: Advancing Multimodal, Long-Context, Reasoning Models. Technical Report / Model Card / Blog Post. https://deepmind. google/models/gemini/pro/, accessed 2025-10-08

    work page 2025

  17. [17]

    Craig Gundersen and James P. Ziliak. 2015. Food insecurity and health outcomes. Health Affairs34, 11 (2015), 1830–1839

    work page 2015

  18. [18]

    Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, et al . 2025. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning.arXiv preprint arXiv:2501.12948(2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  19. [19]

    E ISO. 2010. ISO 9241-2010: 2010-Ergonomics of human-system interaction

    work page 2010

  20. [20]

    Jeongwon Jo, He Zhang, Jie Cai, and Nitesh Goyal. 2025. AI Trust Reshaping Ad- ministrative Burdens: Understanding Trust-Burden Dynamics in LLM-Assisted Benefits Systems. InProceedings of the 2025 ACM Conference on Fairness, Account- ability, and Transparency. 1172–1183

    work page 2025

  21. [21]

    Dean Jolliffe et al. 2024. Food stamps and America’s poorest.American Journal of Agricultural Economics106, 4 (2024), 1380–1409

    work page 2024

  22. [22]

    Rui Maia and Joao C Ferreira. 2018. Context-aware food recommendation system. Context-aware food recommendation system(2018), 349–356

    work page 2018

  23. [23]

    Behzad Maleki Vishkaei and Pietro De Giovanni. 2024. Smart food-sharing platforms for social sustainability: a heuristic algorithm approach.International Transactions in Operational Research(2024)

    work page 2024

  24. [24]

    Manus. n.d.. Manus App. https://manus.im/app Accessed October 8, 2025

    work page 2025

  25. [25]

    Martin, Elizabeth Maddocks, Yiqiang Chen, Stephen E

    Molly S. Martin, Elizabeth Maddocks, Yiqiang Chen, Stephen E. Gilman, and Ian Colman. 2016. Food insecurity and mental illness: disproportionate impacts in the context of perceived stress and social isolation.Public Health132 (2016), 86–91

    work page 2016

  26. [26]

    Bartoo, and John H

    Lynn McIntyre, Allan C. Bartoo, and John H. Emery. 2014. When working is not enough: food insecurity in the Canadian labor force.Public Health Nutrition17, 1 (2014), 49–57

    work page 2014

  27. [27]

    Urquia, and Valerie Tarasuk

    Fei Men, Benedikt Fischer, Marcelo L. Urquia, and Valerie Tarasuk. 2021. Food insecurity, chronic pain, and use of prescription opioids.SSM-Population Health 14 (2021), 100768

    work page 2021

  28. [28]

    Meta AI. 2024. Introducing Llama 3.1: Our Most Capable Models to Date. https: //ai.meta.com/blog/introducing-llama-3-1. Accessed: 2025-10-08

    work page 2024

  29. [29]

    Alexis Millerschultz, Lawton Lanier Nalley, Brandon McFadden, Rodolfo Nayga, and Wei Yang. 2025. Required informational barriers to accessing groceries from food banks.Food Security17, 1 (2025), 9–25

    work page 2025

  30. [30]

    MiniMax. n.d.. MiniMax Agent. https://agent.minimaxi.com/ Accessed October 8, 2025

    work page 2025

  31. [31]

    OpenAI. 2025. ChatGPT (GPT-5/4o). https://chat.openai.com/ Large language model, accessed October 8, 2025

    work page 2025

  32. [32]

    OpenAI. 2025. gpt-oss-120b & gpt-oss-20b Model Card. arXiv:2508.10925 [cs.CL] https://arxiv.org/abs/2508.10925

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  33. [33]

    Qualtrics, LLC. 2024. Qualtrics API Documentation. https://api.qualtrics.com/. Accessed: 2025-10-07

    work page 2024

  34. [34]

    Harmony Rhoades, Suzanne L Wenzel, Eric Rice, Hailey Winetrobe, and Benjamin Henwood. 2017. No digital divide? Technology use among homeless adults. Journal of Social Distress and the Homeless26, 1 (2017), 73–77

    work page 2017

  35. [35]

    Dorothy Rosenbaum. 2013. SNAP is Effective and Efficient.Center on Budget and Policy Priorities(2013)

    work page 2013

  36. [36]

    Amy Rosenthal and Kathe Newman. 2019. Beyond the shadow state: The pub- lic–private food assistance system as networked governance.Urban Affairs Review55, 5 (2019), 1433–1455

    work page 2019

  37. [37]

    Marcus Sammer, Kijin Seong, Norma Olvera, Susie L Gronseth, Elizabeth Anderson-Fletcher, Junfeng Jiao, Alison Reese, and Ioannis A Kakadiaris. 2024. AI-FEED: prototyping an AI-Powered platform for the food charity ecosystem. International Journal of Computational Intelligence Systems17, 1 (2024), 259

    work page 2024

  38. [38]

    Carol Strike, Katherine Rudzinski, Jeffrey Patterson, and Martha Millson. 2012. Frequent food insecurity among injection drug users: correlates and concerns. BMC Public Health12 (2012), 1–9

    work page 2012

  39. [39]

    Infinite Agent

    Flowith Team. 2025. Flowith Neo: The “Infinite Agent” for Long-Term, Multi-Step Tasks. Official Agent Documentation / Blog / Release Note. https://doc.flowith. io/agent-mode-neo/about-agent-neo

    work page 2025

  40. [40]

    Department of Agriculture

    U.S. Department of Agriculture. 2021. Definitions of Food Security. Retrieved from https://www.ers.usda.gov

    work page 2021

  41. [41]

    Weiser, David R

    Sheri D. Weiser, David R. Bangsberg, Susan Kegeles, Kathy Ragland, Margot B. Kushel, and Edward A. Frongillo. 2009. Food insecurity among homeless and marginally housed individuals living with HIV/AIDS in San Francisco.AIDS and Behavior13 (2009), 841–848

    work page 2009

  42. [42]

    Montaner, and Evan Wood

    Dan Werb, Thomas Kerr, Ruirui Zhang, Julio S. Montaner, and Evan Wood. 2010. Methamphetamine use and malnutrition among street-involved youth.Harm Reduction Journal7 (2010), 1–4

    work page 2010

  43. [43]

    Whittle, Lisa A

    Henry J. Whittle, Lisa A. Sheira, Edward A. Frongillo, Kartika Palar, Jennifer Cohen, Debika Merenstein, and Sheri D. Weiser. 2019. Longitudinal associations between food insecurity and substance use in a cohort of women with or at risk for HIV in the United States.Addiction114, 1 (2019), 127–136

    work page 2019

  44. [44]

    An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, et al. 2025. Qwen3 technical report.arXiv preprint arXiv:2505.09388(2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  45. [45]

    Yuanyuan Yang, Ruopeng An, Cao Fang, and Dan Ferris. 2025. Artificial Intelli- gence in Food Bank and Pantry Services: A Systematic Review.Nutrients17, 9 (2025), 1461

    work page 2025

  46. [46]

    Yuehao Yin, Huiyan Qi, Bin Zhu, Jingjing Chen, Yu-Gang Jiang, and Chong-Wah Ngo. 2025. Foodlmm: A versatile food assistant using large multi-modal model. IEEE Transactions on Multimedia(2025)

    work page 2025

  47. [47]

    Zheyuan Zhang, Zehong Wang, Tianyi Ma, Varun Sameer Taneja, Sofia Nelson, Nhi Ha Lan Le, Keerthiram Murugesan, Mingxuan Ju, Nitesh V Chawla, Chuxu Zhang, et al. 2025. Mopi-hfrs: A multi-objective personalized health-aware food recommendation system with llm-enhanced interpretation. InProceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and...

    work page 2025

  48. [48]

    Use- fulness

    Andreas Öttl et al. 2025. Agent-Based Modelling of Food Systems: A Scoping Review.Environmental Modelling & Software(2025). doi:10.1016/j.envsoft.2025. 105230 A Metric Computation To ensure reproducibility and clarity, we provide detailed formula- tions for each metric used in Section??. (1) Food Bank Retrieval.We evaluate the accuracy of geographic retri...

    work page doi:10.1016/j.envsoft.2025 2025