Optimizing the advertiser's cumulative value of winning impressions under budget constraints poses a complex challenge in online advertising, under the paradigm of AI-Generated Bidding (AIGB). Advertisers often have personalized objectives but limited historical interaction data, resulting in few-shot scenarios where traditional reinforcement learning (RL) methods struggle to perform effectively. Large Language Models (LLMs) offer a promising alternative for AIGB by leveraging their in-context learning capabilities to generalize from limited data. However, they lack the numerical precision required for fine-grained optimization. To address this limitation, we introduce GRPO-Adaptive, an efficient LLM post-training strategy that enhances both reasoning and numerical precision by dynamically updating the reference policy during training. Built upon this foundation, we further propose DARA, a novel dual-phase framework that decomposes the decision-making process into two stages: a few-shot reasoner that generates initial plans via in-context prompting, and a fine-grained optimizer that refines these plans using feedback-driven reasoning. This separation allows DARA to combine LLMs' in-context learning strengths with precise adaptability required by AIGB tasks. Extensive experiments on both real-world and synthetic data environments demonstrate that our approach consistently outperforms existing baselines in terms of cumulative advertiser value under budget constraints.
Industrial advertising question answering (QA) is a high-stakes task in which hallucinated content, particularly fabricated URLs, can lead to financial loss, compliance violations, and legal risk. Although Retrieval-Augmented Generation (RAG) is widely adopted, deploying it in production remains challenging because industrial knowledge is inherently relational, frequently updated, and insufficiently aligned with generation objectives. We propose a reinforced co-adaptation framework that jointly optimizes retrieval and generation through two components: (1) Graph-aware Retrieval (GraphRAG), which models entity-relation structure over a high-citation knowledge subgraph for multi-hop, domain-specific evidence selection; and (2) evidence-constrained reinforcement learning via Group Relative Policy Optimization (GRPO) with multi-dimensional rewards covering faithfulness, style compliance, safety, and URL validity. Experiments on an internal advertising QA dataset show consistent gains across expert-judged dimensions including accuracy, completeness, and safety, while reducing the hallucination rate by 72\%. A two-week online A/B test demonstrates a 28.6\% increase in like rate, a 46.2\% decrease in dislike rate, and a 92.7\% reduction in URL hallucination. The system has been running in production for over half a year and has served millions of QA interactions.
Retrieval systems primarily address the challenge of matching user queries with the most relevant advertisements, playing a crucial role in e-commerce search advertising. The diversity of user needs and expressions often produces massive long-tail queries that cannot be matched with merchant bidwords or product titles, which results in some advertisements not being recalled, ultimately harming user experience and search efficiency. Existing query rewriting research focuses on various methods such as query log mining, query-bidword vector matching, or generation-based rewriting. However, these methods often fail to simultaneously optimize the relevance and authenticity of the user's original query and rewrite and maximize the revenue potential of recalled ads. In this paper, we propose a Multi-objective aligned Bidword Generation Model (MoBGM), which is composed of a discriminator, generator, and preference alignment module, to address these challenges. To simultaneously improve the relevance and authenticity of the query and rewrite and maximize the platform revenue, we design a discriminator to optimize these key objectives. Using the feedback signal of the discriminator, we train a multi-objective aligned bidword generator that aims to maximize the combined effect of the three objectives. Extensive offline and online experiments show that our proposed algorithm significantly outperforms the state of the art. After deployment, the algorithm has created huge commercial value for the platform, further verifying its feasibility and robustness.
Arka Dutta, Agrik Majumdar, Sombrata Biswas
et al.
This paper proposes a comprehensive framework for the generation of covert advertisements within Conversational AI systems, along with robust techniques for their detection. It explores how subtle promotional content can be crafted within AI-generated responses and introduces methods to identify and mitigate such covert advertising strategies. For generation (Sub-Task~1), we propose a novel framework that leverages user context and query intent to produce contextually relevant advertisements. We employ advanced prompting strategies and curate paired training data to fine-tune a large language model (LLM) for enhanced stealthiness. For detection (Sub-Task~2), we explore two effective strategies: a fine-tuned CrossEncoder (\texttt{all-mpnet-base-v2}) for direct classification, and a prompt-based reformulation using a fine-tuned \texttt{DeBERTa-v3-base} model. Both approaches rely solely on the response text, ensuring practicality for real-world deployment. Experimental results show high effectiveness in both tasks, achieving a precision of 1.0 and recall of 0.71 for ad generation, and F1-scores ranging from 0.99 to 1.00 for ad detection. These results underscore the potential of our methods to balance persuasive communication with transparency in conversational AI.
Auto-bidding is crucial in facilitating online advertising by automatically providing bids for advertisers. While previous work has made great efforts to model bidding environments for better ad performance, it has limitations in generalizability across environments since these models are typically tailored for specific bidding scenarios. To this end, we approach the scenario-independent principles through a unified function that estimates the achieved effect under specific bids, such as budget consumption, gross merchandise volume (GMV), page views, etc. Then, we propose a bidding foundation model Bid2X to learn this fundamental function from data in various scenarios. Our Bid2X is built over uniform series embeddings that encode heterogeneous data through tailored embedding methods. To capture complex inter-variable and dynamic temporal dependencies in bidding data, we propose two attention mechanisms separately treating embeddings of different variables and embeddings at different times as attention tokens for representation learning. On top of the learned variable and temporal representations, a variable-aware fusion module is used to perform adaptive bidding outcome prediction. To model the unique bidding data distribution, we devise a zero-inflated projection module to incorporate the estimated non-zero probability into its value prediction, which makes up a joint optimization objective containing classification and regression. The objective is proven to converge to the zero-inflated distribution. Our model has been deployed on the ad platform in Taobao, one of the world's largest e-commerce platforms. Offline evaluation on eight datasets exhibits Bid2X's superiority compared to various baselines and its generality across different scenarios. Bid2X increased GMV by 4.65% and ROI by 2.44% in online A/B tests, paving the way for bidding foundation model in computational advertising.
In the realm of online advertising, advertisers partake in ad auctions to obtain advertising slots, frequently taking advantage of auto-bidding tools provided by demand-side platforms. To improve the automation of these bidding systems, we adopt generative models, namely the Decision Transformer (DT), to tackle the difficulties inherent in automated bidding. Applying the Decision Transformer to the auto-bidding task enables a unified approach to sequential modeling, which efficiently overcomes short-sightedness by capturing long-term dependencies between past bidding actions and user behavior. Nevertheless, conventional DT has certain drawbacks: (1) DT necessitates a preset return-to-go (RTG) value before generating actions, which is not inherently produced; (2) The policy learned by DT is restricted by its training data, which is consists of mixed-quality trajectories. To address these challenges, we introduce the R* Decision Transformer (R* DT), developed in a three-step process: (1) R DT: Similar to traditional DT, R DT stores actions based on state and RTG value, as well as memorizing the RTG for a given state using the training set; (2) R^ DT: We forecast the highest value (within the training set) of RTG for a given state, deriving a suboptimal policy based on the current state and the forecasted supreme RTG value; (3) R* DT: Based on R^ DT, we generate trajectories and select those with high rewards (using a simulator) to augment our training dataset. This data enhancement has been shown to improve the RTG of trajectories in the training data and gradually leads the suboptimal policy towards optimality. Comprehensive tests on a publicly available bidding dataset validate the R* DT's efficacy and highlight its superiority when dealing with mixed-quality trajectories.
This paper introduces a signature-based framework for detecting advertising creative fatigue using path signatures, a geometric representation from rough path theory. Creative fatigue -- the degradation of creative effectiveness under repeated exposure -- is operationally important in digital marketing because delayed detection can translate directly into avoidable opportunity cost. We reframe fatigue monitoring as a geometric change detection problem: advertising performance trajectories are embedded as paths and represented by truncated (log-)signatures, enabling detection of changes in trend, volatility, and non-linear dynamics beyond simple mean or variance shifts. We further connect statistical detection to managerial decision-making via an explicit quantification of performance loss relative to a benchmark period. Because proprietary production data cannot be released, we evaluate the proposed framework on a synthetic panel dataset designed to mimic realistic impression volumes and noisy day-to-day CTR dynamics. We define observed CTR as the realised binomial rate $CTR_t := C_t/I_t$ using daily clicks $C_t$ and impressions $I_t$. The accompanying CSV also contains a pre-computed CTR field (e.g., due to rounding or upstream derivation), but all modelling and evaluation in this paper use $C_t/I_t$. Crucially, the dataset does not include injected changepoints; we therefore define an operational ground truth for ``fatigue onset'' based on a noise-robust CTR estimate and a sustained deterioration relative to a recent-best baseline. We report lead-time (early warning) and alert-burden metrics under this operational definition, and provide a sensitivity analysis over the detector's primary tuning parameters. The methodology scales linearly in time-series length for fixed signature depth and is suitable for monitoring large creative portfolios.
Nurlaillah Sari Amallah, Gun Gun Heryanto, Novi Andayani Praptiningsih
et al.
Abstract Media independence is often tested where political actors and advertisers can exert pressure. And we know less about how employee-based share ownership works as a governance mechanism for independence. This study examines how ownership structure conditions newsroom autonomy through a qualitative study of Tempo. Using Giddens’ structuration theory and Mosco’s political economy of communication, we analyse how newsroom rules and resources interact with allocative power from markets and state-linked advertisers. Data come from observation of editorial routines, document analysis of Tempo’s share ownership, and interviews with the Editor-in-Chief, current journalists, and six former journalists. Findings show that dispersed ownership involving employees and foundations strengthens internal bargaining power and supports editorial decisions made in collective meetings, helping the newsroom resist threats and advertising boycotts while maintaining verification norms. At the same time, reliance on major advertisers creates channels for pressure that require continual organisational buffering. The study concludes that employee share ownership can function as a protective governance arrangement for media independence and suggests future research to explore comparative and longitudinal research linking ownership structures to content and observable independence.
Online advertising has become a core revenue driver for the internet industry, with ad auctions playing a crucial role in ensuring platform revenue and advertiser incentives. Traditional auction mechanisms, like GSP, rely on the independent CTR assumption and fail to account for the influence of other displayed items, termed externalities. Recent advancements in learning-based auctions have enhanced the encoding of high-dimensional contextual features. However, existing methods are constrained by the "allocation-after-prediction" design paradigm, which models set-level externalities within candidate ads and fails to consider the sequential context of the final allocation, leading to suboptimal results. This paper introduces the Contextual Generative Auction (CGA), a novel framework that incorporates permutation-level externalities in multi-slot ad auctions. Built on the structure of our theoretically derived optimal solution, CGA decouples the optimization of allocation and payment. We construct an autoregressive generative model for allocation and reformulate the incentive compatibility (IC) constraint into minimizing ex-post regret that supports gradient computation, enabling end-to-end learning of the optimal payment rule. Extensive offline and online experiments demonstrate that CGA significantly enhances platform revenue and CTR compared to existing methods, while effectively approximating the optimal auction with nearly maximal revenue and minimal regret.
In display advertising, advertisers want to achieve a marketing objective with constraints on budget and cost-per-outcome. This is usually formulated as an optimization problem that maximizes the total utility under constraints. The optimization is carried out in an online fashion in the dual space - for an incoming Ad auction, a bid is placed using an optimal bidding formula, assuming optimal values for the dual variables; based on the outcome of the previous auctions, the dual variables are updated in an online fashion. While this approach is theoretically sound, in practice, the dual variables are not optimal from the beginning, but rather converge over time. Specifically, for the cost-constraint, the convergence is asymptotic. As a result, we find that cost-control is ineffective. In this work, we analyse the shortcomings of the optimal bidding formula and propose a modification that deviates from the theoretical derivation. We simulate various practical scenarios and study the cost-control behaviors of the two algorithms. Through a large-scale evaluation on the real-word data, we show that the proposed modification reduces the cost violations by 50%, thereby achieving a better cost-control than the theoretical bidding formula.
We consider a class of optimal control problems that arise in connection with optimal advertising under uncertainty. Two main features appear in the model: a delay in the control variable driving the state dynamics; a mean-field term both in the state dynamics and in the utility functional, taking into account for other agents. We interpret the model in a competitive environment, hence we set it in the framework of Mean Field Games. We rephrase the problem in an infinite dimensional setting, in order to obtain the associated Mean Field Game system. Finally, we specify the problem to a simple case, and solve it providing an explicit solution.
Lucy Atkinson, Dorothy J. Dankel, Katherine D. Romanak
Environmental monitoring at geologic CO2 storage sites is required by regulations for the purposes of environmental protection and emissions accounting in the case of leakage to surface. However, another very important goal of environmental monitoring is to assure stakeholders that the project is monitored for safety and effectiveness. With current efforts to optimize monitoring for cost-effectiveness, the question remains: will optimization of monitoring approaches degrade stakeholder assurance, or do heavily-instrumented sites communicate higher risk to a stakeholder? We report the results of a stakeholder survey in Gulf Coast states of the US where carbon capture and storage (CCS) is developing quickly. We rely on a 2 by 2 factorial experiment in which we manipulate message complexity (complex v. simple) and social norm (support from scientists v. support from community members). Subjects were randomly assigned to one of four conditions: 1) complex message with scientist support; 2) complex message with community member support; 3) simple message with scientist support; or 4) simple message with community member support. In addition to the experimental stimuli, subjects were also asked about their need for cognition, attitudes toward science and scientists, attitudes about climate change and support for carbon capture and storage (CCS). Our sample is drawn from residents in states bordering the western Gulf of Mexico (Texas, Louisiana, Florida) where CO2 geologic storage is being planned both onshore and offshore. The results offer important implications for public outreach efforts to key stakeholders.
Science, General. Including nature conservation, geographical distribution
Importance. Soft skills formation as a part of teaching a foreign language to students of the educational direction “Tourism” contributes to communication skills, empathy, creativity and flexibility of thinking, which are necessary in their future professional activities. The purpose of this study is: 1) analysis of the goals of teaching a foreign language and their correlation with the process of developing hard and soft skills; 2) conducting an empirical study to test the possibility of developing soft skills when teaching a foreign language; 3) identification of significant issues related to the interpretation and use of various cultural codes in communication activities. Attention is focused on the fact that behavioral stereotypes can be variably interpreted depending on which linguistic and cultural communities the communicants belong to. It is emphasized that familiarity with the cultural codes of the studied language and trainings on the development of soft skills can significantly improve the decoding and communicative abilities of students in the process of working with the material for creating advertising texts of tourist destinations and excursion programmes.Materials and Methods. The experiment is carried out at Derzhavin Tambov State University with the aim of compiling on its basis fragments of research based on the relevance of the topic, a review of scientific literature (definitions and experimental data published in this field of knowledge), selection of bibliography and results’ analysis of its experimental verification. The experiment is based on the use of foreign languages teaching methods. The criteria for selecting educational material are texts related to the popularization and advertising of tourist destinations and step-by-step training is aimed at developing soft skills within the framework of game-based educational activities.Results and Discussion. Step-by-step training has shown its success in solving a number of assigned tasks. Students become more flexible in the communication process, showing the correct reactions in accordance with the situation and the target audience, the skills of formulating and implementing communication goals are developed (in such aspects as informational, informative, normative), choosing and using established constructs within the framework of the relevant situation, analysis and synthesis of the results of the dialogue.Conclusion. Based on the results of the experiment, the conclusion is made about the significant didactic potential for soft skills formation within teaching a foreign language. The cognitive and affective aspects of student training have been identified, contributing to the soft skills development. Specific stages of work are proposed, combining various forms of presenting information, professional competence and maintaining successful dialogue with different linguistic and cultural groups in a foreign language.
Zuzana Černeková, Zuzana Berger Haladová, Ján Špirka
et al.
Outdoor advertising, such as roadside billboards, plays a significant role in marketing campaigns but can also be a distraction for drivers, potentially leading to accidents. In this study, we propose a pipeline for evaluating the significance of roadside billboards in videos captured from a driver's perspective. We have collected and annotated a new BillboardLamac dataset, comprising eight videos captured by drivers driving through a predefined path wearing eye-tracking devices. The dataset includes annotations of billboards, including 154 unique IDs and 155 thousand bounding boxes, as well as eye fixation data. We evaluate various object tracking methods in combination with a YOLOv8 detector to identify billboard advertisements with the best approach achieving 38.5 HOTA on BillboardLamac. Additionally, we train a random forest classifier to classify billboards into three classes based on the length of driver fixations achieving 75.8% test accuracy. An analysis of the trained classifier reveals that the duration of billboard visibility, its saliency, and size are the most influential features when assessing billboard significance.
Research has investigated the impact of the COVID-19 pandemic on business performance and survival, indicating particularly adverse effects for small and midsize businesses (SMBs). Yet only limited work has examined whether and how online advertising technology may have helped shape these outcomes, particularly for SMBs. The aim of this study is to address this gap. By constructing and analyzing a novel data set of more than 60,000 businesses in 49 countries, we examine the impact of government lockdowns on business survival. Using discrete-time survival models with instrumental variables and staggered difference-in-differences estimators, we find that government lockdowns increased the likelihood of SMB closure around the world but that use of online advertising technology attenuates this adverse effect. The findings show heterogeneity in country, industry, and business size, which we discuss and is consistent with theoretical expectations.
Chelsy Angelia Timothy, Shinta Shinta, Fransiska Fransiska
et al.
Purpose — This quantitative research aims to analyze and determine the impact of TV advertising and social media influencer endorsements on buying interest in the culinary field.
Method — This research is quantitative in nature, focusing on individuals in Indonesia who utilize and watch digital TV. The research population includes those who meet the specified criteria. A total of 113 respondents participated, and primary data was collected through online questionnaires. The data analysis employed regression techniques.
Result — As a result of this study, both social media influencers and TV advertising have been found to positively and significantly influence buying interest.
Novelty — This research is pioneering in examination of digital TV advertising within the culinary field, and it extends its scope by investigating the widely used advertising medium of social media influencers.
A basic assumption in online advertising is that it is possible to attribute a view of a particular ad creative (i.e., an impression) to a particular web page. In practice, however, the seemingly simple task of ad attribution is challenging due to the scale, complexity and diversity of ad delivery systems. In this paper, we describe a new form of fraud that we call placement laundering, which exploits vulnerabilities in attribution mechanisms. Placement laundering allows malicious actors to inflate revenue by making ad calls that appear to originate from high quality publishers. We describe the basic aspects of placement laundering and give details of two instances found in the wild. One of the instances that we describe abuses the intended functionality of the widely-deployed SafeFrame environment. We describe a placement laundering detection method that is capable of identifying a general class of laundering schemes, and provide results on tests with that method.
This research focuses on the bid optimization problem in the real-time bidding setting for online display advertisements, where an advertiser, or the advertiser's agent, has access to the features of the website visitor and the type of ad slots, to decide the optimal bid prices given a predetermined total advertisement budget. We propose a risk-aware data-driven bid optimization model that maximizes the expected profit for the advertiser by exploiting historical data to design upfront a bidding policy, mapping the type of advertisement opportunity to a bid price, and accounting for the risk of violating the budget constraint during a given period of time. After employing a Lagrangian relaxation, we derive a parametrized closed-form expression for the optimal bidding strategy. Using a real-world dataset, we demonstrate that our risk-averse method can effectively control the risk of overspending the budget while achieving a competitive level of profit compared with the risk-neutral model and a state-of-the-art data-driven risk-aware bidding approach.
Orestis Papakyriakopoulos, Christelle Tessono, Arvind Narayanan
et al.
Online platforms play an increasingly important role in shaping democracy by influencing the distribution of political information to the electorate. In recent years, political campaigns have spent heavily on the platforms' algorithmic tools to target voters with online advertising. While the public interest in understanding how platforms perform the task of shaping the political discourse has never been higher, the efforts of the major platforms to make the necessary disclosures to understand their practices falls woefully short. In this study, we collect and analyze a dataset containing over 800,000 ads and 2.5 million videos about the 2020 U.S. presidential election from Facebook, Google, and TikTok. We conduct the first large scale data analysis of public data to critically evaluate how these platforms amplified or moderated the distribution of political advertisements. We conclude with recommendations for how to improve the disclosures so that the public can hold the platforms and political advertisers accountable.