Gifts: Mechanisms of Social Exchange
- Gifts are transfers of money, goods, or information that create relational obligations, distinguishing them from simple market exchanges.
- They function as mechanisms of reciprocity, signaling, coordination, and influence across digital, social, and algorithmic systems.
- Empirical studies and mathematical models reveal that gifting shapes social organization, drives digital contagion, and underpins targeted marketing and institutional dynamics.
Searching arXiv for the supplied papers and closely related work on gifts.
to=arxiv_search.search 彩神争霸官网 彩神争霸能"query":"all:gifts AND (cat:cs OR cat:econ OR cat:soc-ph OR cat:stat OR cat:physics)","max_results":10,"sort_by":"relevance"}【อ่านข้อความเต็มjson
{"results":[{"arxiv_id":"(Wang et al., 14 Jul 2025)","title":"The Man Behind the Sound: Demystifying Audio Private Attribute Profiling via Multimodal LLM Agents","authors":["Yixuan Li","Huiqiang Zheng","Hanley Zhang","Lichao Sun"],"abstract":"Our research uncovers a novel privacy risk associated with multimodal LLMs (MLLMs): the ability to infer sensitive personal attributes from audio data -- a technique we term audio private attribute profiling. This capability poses a significant threat, as audio can be covertly captured without direct interaction or visibility. Moreover, compared to images and text, audio carries unique characteristics, such as tone and pitch, which can be exploited for more detailed profiling. However, two key challenges exist in understanding MLLM-employed private attribute profiling from audio: (1) the lack of audio benchmark datasets with sensitive attribute annotations and (2) the limited ability of current MLLMs to infer such attributes directly from audio. To address these challenges, we introduce AP2, an audio benchmark dataset that consists of two subsets collected and composed from real-world data, and both are annotated with sensitive attribute labels. Additionally, we propose Gifts, a hybrid multi-agent framework that leverages the complementary strengths of audio-LLMs (ALMs) and LLMs to enhance inference capabilities. Gifts employs an LLM to guide the ALM in inferring sensitive attributes, then forensically analyzes and consolidates the ALM's inferences, overcoming severe hallucinations of existing ALMs in generating long-context responses. Our evaluations demonstrate that Gifts significantly outperforms baseline approaches in inferring sensitive attributes. Finally, we investigate model-level and data-level defense strategies to mitigate the risks of audio private attribute profiling. Our work validates the feasibility of audio-based privacy attacks using MLLMs, highlighting the need for robust defenses, and provides a dataset and framework to facilitate future research.","categories":["cs.CR","cs.SD"]},{"arxiv_id":"(Apagodu et al., 2017)","title":"Analysis of the gift exchange problem","authors":["Erwin Kruyswijk","N. J. A. Sloane","Simon R. Blackburn"],"abstract":"In the gift exchange game there are n players and n wrapped gifts. When a player's number is called, that person can either choose one of the remaining wrapped gifts, or can \"steal\" a gift from someone who has already unwrapped it, subject to the restriction that no gift can be stolen more than a total of sigma times. The problem is to determine the number of ways that the game can be played out, for given values of sigma and n. Formulas and asymptotic expansions are given for these numbers. This work was inspired in part by a 2005 remark by Robert A. Proctor in the On-Line Encyclopedia of Integer Sequences.","categories":["math.CO"]},{"arxiv_id":"(Yuan et al., 2019)","title":"Gift Contagion in Online Groups: Evidence From Virtual Red Packets","authors":["Yi Zeng","Jing Zeng","Qianqian Xie","Bo Zheng","Ting Yan","Khaled Ben Letaifa","Yili Hong"],"abstract":"Gifts are important instruments for forming bonds in interpersonal relationships. Our study analyzes the phenomenon of gift contagion in online groups. Gift contagion encourages social bonds by prompting further gifts; it may also promote group interaction and solidarity. Using data on 36 million online red packet gifts on a large social site in East Asia, we leverage a natural experimental design to identify the social contagion of gift giving in online groups. Our natural experiment is enabled by the randomization of the gift amount allocation algorithm on the platform, which addresses the common challenge of causal identifications in observational data. Our study provides evidence of gift contagion: on average, receiving one additional dollar causes a recipient to send 18 cents back to the group within the subsequent 24 hours. Decomposing this effect, we find that it is mainly driven by the extensive margin -- more recipients are triggered to send red packets. Moreover, we find that this effect is stronger for \"luckiest draw\" recipients, suggesting the presence of a group norm regarding the next red packet sender. Finally, we investigate the moderating effects of group- and individual-level social network characteristics on gift contagion as well as the causal impact of receiving gifts on group network structure. Our study has implications for promoting group dynamics and designing marketing strategies for product adoption.","categories":["cs.CY","cs.SI","q-fin.EC"]},{"arxiv_id":"(Yuan et al., 2017)","title":"Online Red Packets: A Large-scale Empirical Study of Gift Giving on WeChat","authors":["Yujian Zhou","Jingyuan Yang","Cindy Hui","Yuxin He","Zhiyuan Ren","Mingyu Feng","Kunbo Zhang","Ning Hu","Minggao Ouyang","Yuxiao Dong","Jie Tang"],"abstract":"Gift giving is a ubiquitous social phenomenon, and red packets have been used as monetary gifts in Asian countries for thousands of years. In recent years, online red packets have become widespread in China through the WeChat platform. Exploiting a unique dataset consisting of 61 million group red packets and seven million users, we conduct a large-scale, data-driven study to understand the spread of red packets and the effect of red packets on group activity. We find that the cash flows between provinces are largely consistent with provincial GDP rankings, e.g., red packets are sent from users in the south to those in the north. By distinguishing spontaneous from reciprocal red packets, we reveal the behavioral patterns in sending red packets: males, seniors, and people with more in-group friends are more inclined to spontaneously send red packets, while red packets from females, youths, and people with less in-group friends are more reciprocal. Furthermore, we use propensity score matching to study the external effects of red packets on group dynamics. We show that red packets increase group participation and strengthen in-group relationships, which partly explain the benefits and motivations for sending red packets.","categories":["cs.CY","physics.soc-ph","q-fin.EC"]},{"arxiv_id":"(Itao et al., 2022)","title":"Transition of Social Organisations Driven by Gift Relationship","authors":["Takuya Kobayashi","Fuga Chiba","Junya Morita"],"abstract":"Anthropologists have observed gift relationships that establish social relations as well as the transference of goods in many human societies. The totality of such social relations constitutes the network. Social scientists have analysed different types of social organisations with their characteristic networks. However, the factors and mechanisms that cause the transition between these types have hardly been explained. Here, we focus on the gift as the driving force for such changes. We build the model by idealising gift interactions and simulating the consequent social change due to long-term massive interactions. We demonstrate the emergence of disparities and various social organisations depending on the frequency of the gift, consistent with the empirical data. The constructive simulation study, as presented here, explains how people's interactions shape various social structures in response to environmental conditions. Combined with empirical studies, this could contribute to the formulation of a general theory in the social sciences.","categories":["physics.soc-ph"]},{"arxiv_id":"(Gottesman et al., 2013)","title":"Collective Philanthropy: Describing and Modeling the Ecology of Giving","authors":["Russell James III","Benjamin A. M. Mott","Carrie M. Bruneau","John Georghiou","Michael Coghlan","Anna M. Smith","Andrew M. Berdahl","D.B. Larremore","Daniel B. Larremore"],"abstract":"Reflective of income and wealth distributions, philanthropic gifting appears to follow an approximate power-law size distribution as measured by the size of gifts received by individual institutions. We explore the ecology of gifting by analysing data sets of individual gifts for a diverse group of institutions dedicated to education, medicine, art, public support, and religion. We find that the detailed forms of gift-size distributions differ across but are relatively constant within charity categories. We construct a model for how a donor's income affects their giving preferences in different charity categories, offering a mechanistic explanation for variations in institutional gift-size distributions. We discuss how knowledge of gift-sized distributions may be used to assess an institution's gift-giving profile, to help set fundraising goals, and to design an institution-specific giving pyramid.","categories":["physics.soc-ph","q-fin.GN","physics.data-an"]},{"arxiv_id":"(Newham et al., 2022)","title":"The Cost of Influence: How Gifts to Physicians Shape Prescriptions and Drug Costs","authors":["Nicolas Ziebarth","Tímea Laura Molnár"],"abstract":"This paper studies how gifts - monetary or in-kind payments - from drug firms to physicians in the US affect prescriptions and drug costs. We estimate heterogeneous treatment effects by combining physician-level data on antidiabetic prescriptions and payments with causal inference and machine learning methods. We find that payments cause physicians to prescribe more brand drugs, resulting in a cost increase of $30 per dollar received. Responses differ widely across physicians, and are primarily explained by variation in patients' out-of-pocket costs. A gift ban is estimated to decrease drug costs by 3-4%. Taken together, these novel findings reveal how payments shape prescription choices and drive up costs.","categories":["econ.GN","econ.EM","q-fin.GN"]},{"arxiv_id":"(Kong et al., 2024)","title":"Learning to Balance Altruism and Self-interest Based on Empathy in Mixed-Motive Games","authors":["Yunxing Zhang","Chen Gong","Zequn Ma","Jianye Hao","Bolei Zhou","Li Zhao"],"abstract":"Real-world multi-agent scenarios often involve mixed motives, demanding altruistic agents capable of self-protection against potential exploitation. However, existing approaches often struggle to achieve both objectives. In this paper, based on that empathic responses are modulated by inferred social relationships between agents, we propose LASE Learning to balance Altruism and Self-interest based on Empathy), a distributed multi-agent reinforcement learning algorithm that fosters altruistic cooperation through gifting while avoiding exploitation by other agents in mixed-motive games. LASE allocates a portion of its rewards to co-players as gifts, with this allocation adapting dynamically based on the social relationship -- a metric evaluating the friendliness of co-players estimated by counterfactual reasoning. In particular, social relationship measures each co-player by comparing the estimated Q-function of current joint action to a counterfactual baseline which marginalizes the co-player's action, with its action distribution inferred by a perspective-taking module. Comprehensive experiments are performed in spatially and temporally extended mixed-motive games, demonstrating LASE's ability to promote group collaboration without compromising fairness and its capacity to adapt policies to various types of interactive co-players.","categories":["cs.MA","cs.AI"]},{"arxiv_id":"(Kim et al., 16 Jan 2025)","title":"The Spread of Virtual Gifting in Live Streaming: The Case of Twitch","authors":["Jiyeon Kim","Jisu Yoon","Junsol Choi"],"abstract":"This paper examines how gifting spreads among viewers on Twitch, one of the largest live streaming platforms worldwide. Twitch users can give gift subscriptions to other viewers in the chat room, with the majority of gifters opting for community gifting, which is gifting to randomly selected viewers. We identify the random nature of gift-receiving in our data as a natural experiment setting. We investigate whether gift recipients pay it forward, considering various gift types that may either promote or deter the spread of gifting. Our findings reveal that Twitch viewers who receive gift subscriptions are generally more likely to pay it forward than non-recipients, and the positive impact of gift-receiving becomes stronger when the recipient is the sole beneficiary of the giver's gifting behavior. However, we found that gifts from frequent gifters discourage recipients from paying it forward, and gifts from anonymous gifters do not influence the likelihood of viewers becoming future gifters. This research contributes to the existing literature on the spread of online prosocial behavior by providing robust evidence and suggests practical strategies for promoting online gifting.","categories":["cs.CY","econ.GN"]},{"arxiv_id":"(Leone et al., 2016)","title":"Search-and-Rescue Rendezvous","authors":["Steve Alpern","Thomas Lidbetter"],"abstract":"We consider a new type of asymmetric rendezvous search problem in which Agent II needs to give Agent I a gift' which can be in the form of information or material. The gift can either be transfered upon meeting, as in traditional rendezvous, or it can be dropped o? by II at a location he passes, in the hope it will be found by I. The common aim of the two agents is to minimize the time taken for I to either meet II or find the gift. We find optimal agent paths and droppo? times when the search region is a line, the initial distance between the players is known and one or both of the players can leave gifts. When there are no gifts this is the classical asymmetric rendezvous problem solved by Alpern and Gal in 1995 [10]. We exhibit strategies solving these various problems and use arendezvous algorithm' to establish their optimality.","categories":["math.PR","math.OC"]}]}
Searching for additional supplied papers on gift protocols and gift-related retrieval systems.
to=arxiv_search.search 福利彩票天天იწევსjson
{"query":"(\"Conspiracy Santa\" OR \"intent-based product collections\" OR \"physicians shape prescriptions and drug costs\")","max_results":10,"sort_by":"relevance"}ുവനന്തപുരം/json
{"results":[{"arxiv_id":"(Bultel et al., 2020)","title":"A Faster Cryptographer's Conspiracy Santa","authors":["Jules de Terwangne","Olivier Pereira"],"abstract":"In Conspiracy Santa, a variant of Secret Santa, a group of people offer each other Christmas gifts, where each member of the group receives a gift from the other members of the group. To that end, the members of the group form conspiracies, to decide on appropriate gifts, and usually divide the cost of each gift among all participants of that conspiracy. This requires to settle the shared expenses per conspiracy, so Conspiracy Santa can actually be seen as an aggregation of several shared expenses problems. First, we show that the problem of finding a minimal number of transaction when settling shared expenses is NP-complete. Still, there exist good greedy approximations. Second, we present a greedy distributed secure solution to Conspiracy Santa. This solution allows a group of n people to share the expenses for the gifts in such a way that no participant learns the price of his gift, but at the same time notably reduces the number of transactions to 2 × n + 1 with respect to a naïve aggregation of n × (n -- 2). Furthermore, our solution does not require a trusted third party, and can either be implemented physically (the participants are in the same room and exchange money using envelopes) or, over Internet, using a cryptocurrency.","categories":["cs.CR","cs.DC","cs.CY"]},{"arxiv_id":"(Kim et al., 2021)","title":"Intent-based Product Collections for E-commerce using Pretrained LLMs","authors":["Yuya Sasaki","Hiroshi Kasai","Tomoki Taira","Kenji Hatano"],"abstract":"Building a shopping product collection has been primarily a human job. With the manual efforts of craftsmanship, experts collect related but diverse products with common shopping intent that are effective when displayed together, e.g., backpacks, laptop bags, and messenger bags for freshman bag gifts. Automatically constructing a collection requires an ML system to learn a complex relationship between the customer's intent and the product's attributes. However, there have been challenging points, such as 1) long and complicated intent sentences, 2) rich and diverse product attributes, and 3) a huge semantic gap between them, making the problem difficult. In this paper, we use a pretrained LLM (PLM) that leverages textual attributes of web-scale products to make intent-based product collections. Specifically, we train a BERT with triplet loss by setting an intent sentence to an anchor and corresponding products to positive examples. Also, we improve the performance of the model by search-based negative sampling and category-wise positive pair augmentation. Our model significantly outperforms the search-based baseline model for intent-based product matching in offline evaluations. Furthermore, online experimental results on our e-commerce platform show that the PLM-based method can construct collections of products with increased CTR, CVR, and order-diversity compared to expert-crafted collections.","categories":["cs.IR","cs.LG"]},{"arxiv_id":"(Newham et al., 2022)","title":"The Cost of Influence: How Gifts to Physicians Shape Prescriptions and Drug Costs","authors":["Nicolas Ziebarth","Tímea Laura Molnár"],"abstract":"This paper studies how gifts - monetary or in-kind payments - from drug firms to physicians in the US affect prescriptions and drug costs. We estimate heterogeneous treatment effects by combining physician-level data on antidiabetic prescriptions and payments with causal inference and machine learning methods. We find that payments cause physicians to prescribe more brand drugs, resulting in a cost increase of $30 per dollar received. Responses differ widely across physicians, and are primarily explained by variation in patients' out-of-pocket costs. A gift ban is estimated to decrease drug costs by 3-4%. Taken together, these novel findings reveal how payments shape prescription choices and drive up costs.","categories":["econ.GN","econ.EM","q-fin.GN"]}]}
Gifts are transfers of money, goods, information, or reward shares whose analytic importance lies not only in resource movement but also in the relations they create, reinforce, or strategically manipulate. Across recent work in social science, economics, operations research, combinatorics, cryptography, and AI, gifts are modeled as mechanisms of reciprocity, signaling, coordination, redistribution, and influence; as empirical objects with measurable contagion and distributional structure; and, in some technical settings, as formal primitives inside optimization or learning systems (Yuan et al., 2017, Itao et al., 2022, Leone et al., 2016, Kong et al., 2024).
1. Conceptual foundations
A gift relationship is modeled as an interaction in which a donor transfers wealth, goods, prestige items, or even people to a recipient, and the transfer creates obligations “to give, to receive, and to reciprocate.” In this formulation, a gift is neither an impersonal market exchange with immediate equivalence nor pure altruism; it is an open-ended social relation whose balance may remain symmetric or become asymmetric through debt and dependence (Itao et al., 2022).
The digital red packet literature makes the same point in a more operational form. Group red packets on WeChat are treated as gifts because they transfer resources voluntarily, create and reinforce obligations of reciprocity, signal generosity and “mianzi” (face), and maintain relationships in a visible group setting. Senders choose a total amount and number of shares, and recipients claim them on a first-come, first-served basis; in random-split packets, the largest share is marked “Luckiest Draw,” making the act public and socially legible (Yuan et al., 2017).
Research in search theory broadens the concept further. In asymmetric rendezvous on the line, a “gift” may be a water bottle, a supply cache, or important information dropped at a location so that the other agent can find it later; the game ends when the seeker either meets the other agent or finds the gift (Leone et al., 2016). This formalization is significant because it isolates a property common to otherwise different gift systems: a gift can substitute for direct encounter by materializing obligation or assistance in space and time.
A common misconception is that gifts are analytically equivalent to simple transfers. The literature instead treats them as relation-bearing acts. This suggests that the minimal definition of a gift in research is not its object class, but its capacity to bind future action.
2. Gifts, networks, and social organization
A major line of work treats gifts as the micro-foundation of large-scale social structure. In the model of gift-driven social transition, individuals hold wealth and directed tie weights , and each cycle contains a gift stage, a production stage, reciprocation with interest , and network updating by . Wealth production is concave, , and failed reciprocation generates debt, during which debtors cannot initiate new gifts (Itao et al., 2022).
The simulated consequence is a phase structure. Low gift frequency produces band-like, kin-clustered, low-inequality networks; intermediate frequency yields tribe-like formations with stronger economic disparity than social disparity; high gift frequency and/or high reciprocation burden produces chiefdom-like hierarchy, with wealth Gini rising before connectivity Gini . In the hierarchical regime, wealth distributions develop power-law tails with exponent at and 0 at 1, while connectivity also becomes heavy-tailed. SCCS-based empirical indices in the same study show economic and social disparity increasing with “gift degree,” consistent with the model’s claim that repeated gift obligations can transform organizational form (Itao et al., 2022).
An allied empirical literature studies gifts at the institutional level. Philanthropic gift-size distributions are modeled by Zipf and frequency laws, 2 and 3, with 4. Across institutions, the estimated 5 differs systematically by category but remains relatively stable within category: higher education institutions are around 6, health care around 7, combined-purpose charities around 8, and a corrected religious case around 9 (Gottesman et al., 2013).
These exponents matter substantively. Lower 0 implies a heavier upper tail and stronger concentration in top donors; higher 1 implies broader support and less dominance by very large gifts. The ecological model in that literature links institutional exponents to how donors of different incomes rank causes in their own giving portfolios (Gottesman et al., 2013). Taken together with the network model, this indicates that gifts are simultaneously cohesive and stratifying: they bind collectivities, but they also sort actors into central and peripheral positions.
3. Digital gifts and contagion
Digital gift systems make gift processes directly observable at scale. A large WeChat study analyzes 61,501,862 group red packets, 218,301,860 receipt events, and 7,380,110 unique users from 367,361 filtered groups. It distinguishes reciprocal packets, defined by 2 seconds plus recent receipt by the sender, from spontaneous packets, defined by 3 hours. Males, seniors, southerners, and users with more in-group friends are more inclined to send spontaneous red packets; red packets from females, youths, and people with less in-group friends are more reciprocal. At the macro level, net flows run largely east-to-west and south-to-north; in interprovincial regressions, a 1,000 RMB increase in PPP difference is associated with a 9.02 RMB increase in net flow, while each additional kilometer is associated with a 0.574 RMB decrease (Yuan et al., 2017).
That study also estimates causal effects of gift size on group dynamics using propensity score matching. A large spontaneous packet is defined as at least 100 RMB. Relative to matched smaller packets, it causes 0.295 more subsequent senders in the session, 0.415 more following red packets, 52.632 RMB more subsequent session money, and 0.009 additional new in-group friends for the sender within one day (Yuan et al., 2017). The paper’s interpretation is that digital gifts increase participation and strengthen ties.
A second red-packet study exploits the platform’s random allocation algorithm as a natural experiment. On average, receiving one additional dollar causes a recipient to send 18 cents back to the group within the subsequent 24 hours. The effect is driven mainly by the extensive margin, meaning that more recipients are triggered to send red packets, rather than existing senders merely increasing amounts. Direct reciprocity is smaller than generalized reciprocity: about 0.031 goes back to the original sender and about 0.104 to other group members. The effect is stronger for “Luckiest Draw” recipients, and receiving gifts also increases within-group edge formation, with 100 CNY generating about 0.055 additional within-group friends over seven days (Yuan et al., 2019).
Live-streaming gifting exhibits related but not identical dynamics. On Twitch, 65,732 gift subscriptions across 15,653 gifting events are observed in a sample of 956,328 unique users and 11,272,948 chat messages. Because community gifting is randomly allocated among eligible viewers, the setting functions as a natural experiment. Gift recipients are 2.732 times more likely to become future gifters than comparable non-recipients; the positive effect is amplified when the recipient is the sole beneficiary of the event, with an interaction odds ratio of 1.728. However, gifts from frequent gifters dampen pay-it-forward behavior, with an interaction odds ratio of 0.571, and gifts from anonymous gifters show no significant effect (Kim et al., 16 Jan 2025).
These results suggest that digital gifts are not merely virtualized versions of offline presents. Platform rules such as random allocation, public visibility, recipient labeling, and bulk-versus-individual gifting modulate whether gifts propagate as generalized reciprocity, focal norms, or status competition.
4. Institutional gifts, markets, and contested influence
Not all gift systems are solidaristic. In health economics, gifts from pharmaceutical firms to physicians are modeled as monetary and in-kind payments that alter prescribing behavior. Using Medicare Part D and Open Payments data for antidiabetic drugs in Maine and New Hampshire, with Vermont as a ban environment, the study finds that about 16% of physicians receive at least one antidiabetic-related payment in a year; 98% of payments are in-kind, mostly meals; the median paid physician receives 3 payments per year totaling about \$30 (Newham et al., 2022).
Under double machine learning and causal forest estimation, a \$p_{ij}$410 across physicians, and the strongest predictor of responsiveness is patient out-of-pocket exposure, especially the share of Low-Income Subsidy patients. Aggregating the induced claims, the paper estimates that each \$p_{ij}$530 in additional diabetes drug spending, and that a gift ban would reduce diabetes drug costs by about 3–4% (Newham et al., 2022).
This literature is important because it rebuts a naïve altruistic reading of all gifts. Here the gift is a marketing instrument whose function is not mutual bonding but targeted influence. The result does not negate other gift theories; rather, it locates gifting on a broader continuum that includes reciprocity, prestige, coalition-building, and strategic behavioral steering.
Philanthropic distributions reinforce the same point from another angle. Institutions with low $p_{ij}$6 depend disproportionately on elite donors, while those with higher $p_{ij}$7 depend more on broad bases of smaller gifts (Gottesman et al., 2013). A plausible implication is that “gift” is not a morally homogeneous category: identical transfer forms can sustain solidarity, hierarchy, or influence depending on surrounding institutions.
5. Formalization in mathematics, search, and cryptography
In combinatorics, the gift exchange game is formalized as follows: $p_{ij}$8 players sequentially choose either a wrapped gift or an already unwrapped one, subject to the rule that no gift can be stolen more than a total of $p_{ij}$9 times. If $r$0 denotes the number of full scenarios and pool order is fixed, then $r$1, where
$r$2
and $r$3 is a $r$4-restricted Stirling number of the second kind. For $r$5, $r$6 equals the Bessel polynomial value $r$7 and satisfies
$r$8
with asymptotic behavior
$r$9
The paper thereby turns a party game into a family of exact enumerative results (Apagodu et al., 2017).
Operations research treats gifts as search resources. In asymmetric rendezvous on the real line, the classical no-gift value is $\eta = 0.03$0. When one player may leave a gift, the optimal expected termination time drops to $\eta = 0.03$1 with a drop time $\eta = 0.03$2; with two gifts and an “or” stopping rule, the value becomes $\eta = 0.03$3 with both gifts dropped at $\eta = 0.03$4; with two gifts and a stricter “and” requirement, the value is $\eta = 0.03$5 (Leone et al., 2016). The analytical significance is that gifts create additional stopping targets, converting a pure meeting problem into a mixed meeting-or-discovery problem.
Cryptographic protocol design treats gifts as privacy-sensitive shared-expense objects. In Conspiracy Santa, each participant receives a gift funded by all other participants, which induces an aggregation of shared-expense problems. Deciding whether reimbursements can be settled in fewer than $\eta = 0.03$6 transactions is NP-complete. Nevertheless, the paper gives distributed, privacy-preserving protocols requiring 3$\eta = 0.03$7 or $\eta = 0.03$8 transactions, no trusted third party, and either physical-envelope or cryptocurrency implementation (Bultel et al., 2020).
These formalizations show that gifting is not only a social-scientific category. It is also a mathematically tractable structure: a countable game tree, an optimal-control primitive, or a secure multi-party settlement problem.
6. Algorithmic incentives, retrieval systems, and privacy reconfigurations
In multi-agent reinforcement learning, gifts become explicit, learned reward reallocations. LASE defines a gifting weight vector $\eta = 0.03$9 for agent $w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$0, where $w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$1 is the gift from $w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$2 to $w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$3, $w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$4, and total post-gifting reward is
$w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$5
The weights are determined by an empathy-inspired social relationship metric computed from counterfactual $w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$6-values and a perspective-taking module. Negative inferred relationships are clipped to zero, and gifts remain zero-sum. Empirically, two LASE agents in IPD converge to cooperation probability $w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$7; in Cleanup, fairness after gifting reaches $w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$8 versus $w_i^{t\ast\ast} = w_i^{t\ast} + \frac{1+\log(1+w_i^{t\ast})}{100}$9 (Kong et al., 2024). Here gifting is neither metaphor nor exogenous reward shaping; it is an endogenous incentive channel balancing altruism and self-protection.
In information retrieval, gifts appear as semantic shopping intents. Intent-based product collection uses SBERT with triplet loss, BM25-based hard negatives, and category-wise augmentation to match free-form queries such as “freshman bag gifts” to products. Offline, BM25 achieves Recall@100 $G_w$0, whereas the best SBERT variants reach about 0.52–0.53. Online, the PLM-based system yields +16% CTR, +29% CVR, and +60% order-diversity relative to expert-crafted collections (Kim et al., 2021). In this setting, the gift is not the transfer itself but a structured intent condition over recipient, occasion, budget, and style.
A recent privacy paper reuses the term in a different way. “Gifts” is an acronym for Guidance, inference, forensics, scrutinization, consolidation, a hybrid ALM–LLM framework for audio private attribute profiling. On the AP$G_w$1-Com benchmark, Gifts reaches 86.7 ± 1.3 average profiling accuracy; in a smaller human comparison, humans score 67.3 while Gifts scores 89.8. Two defenses sharply reduce performance: in-context unlearning lowers Gifts to 58.3 on AP$G_w$2-Com, and anti-eavesdropping jamming lowers it to 46.2 (Wang et al., 14 Jul 2025). This is a terminological migration rather than a gift-transfer theory, but it is notable because it shows the semantic reach of “gift” in current AI literature and foregrounds privacy risks in audio-mediated inference.
Across these computational settings, gifting functions as incentive design, semantic retrieval target, or even acronymic architecture. This suggests that the term has become a productive abstraction for any structured mechanism that redistributes value, focus, or inferential leverage under constraints.