preliminary desriptions and babble

paper/src/bib/references.bib

@@ -0,0 +1,62 @@
+@techReport{Imperva2025,
+   author = {Imperva},
+   title = {The Rapid Rise of Bots and the Unseen Risk for Business #2025BADBOTREPORT},
+   year = {2025}
+}
+
+@techReport{Xie,
+   abstract = {Autonomous agents that accomplish complex computer tasks with minimal human interventions have the potential to transform human-computer interaction, significantly enhancing accessibility and productivity. However, existing benchmarks either lack an interactive environment or are limited to environments specific to certain applications or domains, failing to reflect the diverse and complex nature of real-world computer use, thereby limiting the scope of tasks and agent scalability. To address this issue, we introduce OSWORLD, the first-of-its-kind scalable, real computer environment for multimodal agents, supporting task setup, execution-based evaluation, and interactive learning across various operating systems such as Ubuntu, Windows, and macOS. OSWORLD can serve as a unified, integrated computer environment for assessing open-ended computer tasks that involve arbitrary applications. Building upon OSWORLD, we create a benchmark of 369 computer tasks involving real web and desktop apps in open domains, OS file I/O, and workflows spanning multiple applications. Each task example is derived from real-world computer use cases and includes a detailed initial state setup configuration and a custom execution-based evaluation script for reliable, reproducible evaluation. Extensive evaluation of state-of-the-art LLM/VLM-based agents on OSWORLD reveals significant deficiencies in their ability to serve as computer assistants. While humans can accomplish over 72.36% of the tasks, the best model achieves only 12.24% success, primarily struggling with GUI grounding and operational knowledge. Comprehensive analysis using OSWORLD provides valuable insights for developing multimodal generalist agents that were not possible with previous benchmarks. Our code, environment, baseline models, and data are publicly available at https://os-world.github.io.},
+   author = {Tianbao Xie and Danyang Zhang and Jixuan Chen and Xiaochuan Li and Siheng Zhao and Ruisheng Cao and Toh Jing Hua and Zhoujun Cheng and Dongchan Shin and Fangyu Lei and Yitao Liu and Yiheng Xu and Shuyan Zhou and Silvio Savarese and Caiming Xiong and Victor Zhong and Tao Yu},
+   title = {OSWORLD: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments},
+   url = {https://os-world.github.io}
+}
+
+@techReport{MarkNtelAdvisors2025,
+   author = {MarkNtel Advisors},
+   institution = {MarkNtel Advisors},
+   title = {Global AI Agent Market Research Report: Forecast (2026–2032)},
+   url = {https://www.marknteladvisors.com/research-library/ai-agent-market.html},
+   year = {2025}
+}
+
+@article{Xia2025,
+   abstract = {Large Language Models (LLMs) have enabled the emergence of LLM agents, systems capable of pursuing under-specified goals and adapting after deployment. Evaluating such agents is challenging because their behavior is open ended, probabilistic, and shaped by system-level interactions over time. Traditional evaluation methods, built around fixed benchmarks and static test suites, fail to capture emergent behaviors or support continuous adaptation across the lifecycle. To ground a more systematic approach, we conduct a multivocal literature review (MLR) synthesizing academic and industrial evaluation practices. The findings directly inform two empirically derived artifacts: a process model and a reference architecture that embed evaluation as a continuous, governing function rather than a terminal checkpoint. Together they constitute the evaluation-driven development and operations (EDDOps) approach, which unifies offline (development-time) and online (runtime) evaluation within a closed feedback loop. By making evaluation evidence drive both runtime adaptation and governed redevelopment, EDDOps supports safer, more traceable evolution of LLM agents aligned with changing objectives, user needs, and governance constraints.},
+   author = {Boming Xia and Qinghua Lu and Liming Zhu and Zhenchang Xing and Dehai Zhao and Hao Zhang},
+   month = {11},
+   title = {Evaluation-Driven Development and Operations of LLM Agents: A Process Model and Reference Architecture},
+   url = {http://arxiv.org/abs/2411.13768},
+   year = {2025}
+}
+
+
+@misc{AmazonvsPerplexity,
+   author = {Shirin Ghaffary and Matt Day},
+   note = {Updated 2025-11-05},
+   title = {Amazon Sues to Stop Perplexity From Using AI Tool to Buy Stuff},
+   url = {https://www.bloomberg.com/news/articles/2025-11-04/amazon-demands-perplexity-stop-ai-agent-from-making-purchases}
+}
+
+
+@inproceedings{Mueller2019,
+   author = {Jonas W Mueller and Vasilis Syrgkanis and Matt Taddy},
+   booktitle = {Advances in Neural Information Processing Systems 32 (NeurIPS 2019)},
+   pages = {15442-15452},
+   title = {Low-Rank Bandit Methods for High-Dimensional Dynamic Pricing},
+   url = {https://proceedings.neurips.cc/paper/2019/file/0a3df70393993583a13c0dd6686f3f32-Paper.pdf},
+   year = {2019}
+}
+
+@article{Amjad2017,
+   abstract = { In this paper, the question of interest is estimating true demand of a product at a given store location and time period in the retail environment based on a single noisy and potentially censored observation. To address this question, we introduce a %non-parametric framework to make inference from multiple time series. Somewhat surprisingly, we establish that the algorithm introduced for the purpose of "matrix completion" can be used to solve the relevant inference problem. Specifically, using the Universal Singular Value Thresholding (USVT) algorithm [7], we show that our estimator is consistent: the average mean squared error of the estimated average demand with respect to the true average demand goes to 0 as the number of store locations and time intervals increase to $\infty$. We establish naturally appealing properties of the resulting estimator both analytically as well as through a sequence of instructive simulations. Using a real dataset in retail (Walmart), we argue for the practical relevance of our approach. },
+   author = {Muhammad J. Amjad and Devavrat Shah},
+   doi = {10.1145/3154489},
+   issue = {2},
+   journal = {Proceedings of the ACM on Measurement and Analysis of Computing Systems},
+   month = {12},
+   pages = {1-28},
+   publisher = {Association for Computing Machinery (ACM)},
+   title = {Censored Demand Estimation in Retail},
+   volume = {1},
+   url = {https://par.nsf.gov/servlets/purl/10066022},
+   year = {2017}
+}

paper/src/chapters/01-intro.tex

@@ -8,9 +8,22 @@
 
 \section{Introduction}
 
-Research Objectives and Contribution: What are we making, why and who should care?
+In this paper we present an exploration and defense against the persence of new comercial entities present in digitially powered platforms. This research aims to establish the following contributions: definition and formalization of the existence of non-human transactors in e-commerce platforms, development of a testing-ground for capturing the behavioral essence off these transactors on a large variety of digital systems, construction of a discriminative model (to prove separability) as a strong learner for downstream mitigation of contamination by non-human entities, translation of such learned separability into existing dynamic pricing machine learning loops, finally we establish a high-level KPI affecting causal effect and cost-saving framework for the future of commerce done on the internet with the presence of such non-human learners.
+
+This research effort touches a large variety of domains, spanning those such as: behavioral economics for understanding the rationality of behavior as theorizied by the concept of homo economicus, agent-based modeling in our effort to translate our learned separability into disjoint dynamic pricing systems, reinforcement learning which serves as the SOTA for price-learners, dynamic pricing and economics market theory of equilibrium to understand the risks of possible supra-competitive pricing phenomena in cases of adversarial pricing systems (driving the market out of equilibrium).
 
 \subsection{Motivation and Market Context}
-Current market dynamics and trends of dynamic pricing and AI agents. Future projections of AI agents. Key stakeholders that are discussing this and reporting on it (Thales). Who is most affected
+
+The current innovation boom in generative artificial inteligence and its applications to knowledge-based work tasks, has brought many competing technologies for browser-use automation, with most benchmarks and evals motivating \cite{Xia2025} the development of ability focusing on the performance of commercial research, understanding and transaction execution. \cite{Xie} The ``AI Agent'' market is forecasted to grow from around USD 5-8 billion in 2025 to USD 42-52 billion by 2030. This surge reflects adoption in e-commerce, customer service, and enterprise automation, where agents handle interactions previously done by humans. Begging the question of how should these systems be designed for future robustness aswell as how to maintain a competitive edge in the analytical components of these ecommerce platforms. \cite{MarkNtelAdvisors2025}
+
+The key stakeholders of the threat presented by the shift towards a growing share of the traffic coming from agents include online businesses and platform operators (especially in bot-heavy sectors like retail, travel, and financial services), their security/fraud and engineering teams, end users whose accounts and data are exposed (and whose experience degrades), regulators or legal stakeholders responding to breaches and fraud, and the attackers/bot operators driving the automation.\cite{Imperva2025}
+
+The industry has already seen legal action coming from large cases like Amazon against Perplexity \cite{AmazonvsPerplexity} which comes from the nature of the difficulty of identifying traffic coming from hybrid systems like the Commet browser, which is a system that this paper does explore in order to better understand the nature of how the interaction data looks like and what it means for dynamic pricing or recommendation systems down the line. This observed impact indicates a need for prevention of seconddary negative effects on the ``legacy'' systems which power modern revenue sources for many companies. Dynamic pricing algorithms rely on directly trnslating demand features $q$ to new price assignments $\hat{p}$ across a catalogue of products of size $N$.
+
 \subsection{Solution Space Overview}
+The dynamic pricing systems, as presented in \cite{Mueller2019} often deal with sparse low-rank data of demand signals, which in combination with contamination from agents makes for a very complex set of interactions which have an impact on the pricing. To further complicate the problem in certain comercial settings, such as the one presented in \cite{Amjad2017} must addreses the true demand of products under censored observations. This sets us up with a good formulation for handling demand in our case of multiple kinds of comercial mediators  $\hat{q} \gets q_A + q_H$ where $A$ represents the distribution of demand generated by agentic mediators and $H$ represents that of true human demand.
+
+The previously described goal of
+
+
 Different approaches and perspectives, here also add a preview of what will be developed and explored in the lit review.

paper/src/chapters/03-methodology.tex

@@ -15,6 +15,15 @@ We gether interaction data from users interacting with a sample platform simulat
 Mathematical demonstration and validation of the COI and citation backed evidence, and framework overview + show harm to user via other cost distortions. Maybe split into 3.2.1 (COI Theory) and 3.2.2 (Framework Design)
 
 \subsection{System Architecture}
+
+\begin{figure}[ht]
+  \resizebox{\columnwidth}{!}{%
+    \input{chapters/loop_figure.tex}
+  }
+  \caption{Overview of the Dynamic Pricing Tasks.}
+\end{figure}
+
+
 \begin{figure}[ht]
 \centering
 \begin{tikzpicture}[