Contributions in Merlin

In recent years, a surge of criminal activities with cross-cryptocurrency trades have emerged in Ethereum, the second-largest public blockchain platform. Most of the existing anomaly detection methods utilize the traditional machine learning with feature engineering or graph representation learning technique to capture the information in transaction network. However, these methods either ignore the timestamp information and the transaction flow direction information in transaction network or only consider single transaction network, the cross-cryptocurrency trading patterns in Ethereum are usually ignored. In this paper, we introduce a Multi-layer Temporal Transaction Anomaly Detection (MT$^2$AD) model in Ethereum network with graph neural network. Specifically, for a given Ethereum token transaction network, we first extract its initial features including the structure subgraph and edge’s feature. Then, we model the temporal information in subgraph as a series of network snapshots according to the timestamp on each edge and time window. To capture the cross-cryptocurrency trading patterns, we combine the snapshots from multiple token transactions at a given timestamp, and we consider it as a new combined graph. We further use the graph convolution encoder with attention mechanism and pooling operation on this new graph to obtain the graph-level embedding, and we transform the anomaly detection on dynamic multi-layer Ethereum transaction networks as a graph classification task with these graph-level embeddings. MT$^2$AD can integrate the transaction structure feature, edge’s feature and cross-cryptocurrency trading patterns into a framework to perform anomaly detection with graph neural networks. Experiments on three real-world multi-layer transaction networks show that the proposed MT$^2$AD (0.8789 Precision, 0.9375 Recall, 0.4987 FbMacro and 0.9351 FbWeighted) can achieve the best performance on most evaluation metrics in comparison with some competing approaches, and the effectiveness in consideration of multiple tokens is also demonstrated.

We use BERT, an AI-based algorithm for language understanding, to quantify regulatory climate risk disclosures and analyze their impact on the term structure in the credit default swap (CDS) market. Risk disclosures can either increase or decrease CDS spreads, depending on whether the disclosure reveals new risks or reduces uncertainty. Training BERT to differentiate between transition and physical climate risks, we find that disclosing transition risks increases CDS spreads after the Paris Climate Agreement of 2015, while disclosing physical risks decreases the spreads. In addition, we also find that the election of Trump had a negative impact on CDS spreads for firms exposed to transition risk. These impacts are consistent with theoretical predictions and economically and statistically significant.

Over fifty countries have introduced regulatory sandboxes to foster financial innovation. This article conducts the first evaluation of their ability to improve fintechs’ access to capital and attendant real effects. Exploiting the staggered introduction of the UK sandbox, we establish that firms entering the sandbox see an increase of 15% in capital raised post-entry. Their probability of raising capital increases by 50%. Sandbox entry also has a significant positive effect on survival rates and patenting. Investigating the mechanism, we present evidence consistent with lower asymmetric information and regulatory costs.

While the crypto market has experienced volatility and skepticism, the underlying blockchain technology has continually evolved since the introduction of Bitcoin in 2009. Though Bitcoin has doubled since last year, the focus has shifted to non-fungible tokens (NFTs) and infrastructure protocols like Chainlink and Graph.

Vision Transformers (ViTs) have shown impressive performance in computer vision, but their high computational cost, quadratic in the number of tokens, limits their adoption in computation-constrained applications. However, this large number of tokens may not be necessary, as not all tokens are equally important. In this paper, we investigate token pruning to accelerate inference for object detection and instance segmentation, extending prior works from image classification. Through extensive experiments, we offer four insights for dense tasks: (i) tokens should not be completely pruned and discarded, but rather preserved in the feature maps for later use. (ii) reactivating previously pruned tokens can further enhance model performance. (iii) a dynamic pruning rate based on images is better than a fixed pruning rate. (iv) a lightweight, 2-layer MLP can effectively prune tokens, achieving accuracy comparable with complex gating networks with a simpler design. We evaluate the impact of these design choices on COCO dataset and present a method integrating these insights that outperforms prior art token pruning models, significantly reducing performance drop from ~1.5 mAP to ~0.3 mAP for both boxes and masks. Compared to the dense counterpart that uses all tokens, our method achieves up to 34% faster inference speed for the whole network and 46% for the backbone.

It is widely assumed that blockchain should, in principle, lead to decentralization. Yet, in practice, many enterprise blockchains are highly centralized. To explain this conundrum, we conduct a multi-case study of four enterprise blockchains: Walmart DL Freight, Contour, Chronicled MediLedger, and Cardossier. Exploring the dynamics of participant dominance and platform openness during their formative stages, we theorize that these blockchains correspond to the distinct archetypes of Chief, Clan, Custodian, and Consortium, respectively. Importantly, these archetypes shape the subsequent evolution of the governance approach, thus explaining why and how enterprise blockchains with dominant participants and limited openness later exhibit more centralized governance.

Algorithms take decisions that affect humans, and have been shown to perpetuate biases and discrimination. Decisions by algorithms are subject to different interpretations. Algorithms’ behaviors are basis for the construal of moral assessment and standards. Yet we lack an understanding of how algorithms impact on social construction processes, and vice versa. Without such understanding, social construction processes may be disrupted and, eventually, may impede moral progress in society. We analyze the public discourse that emerged after a significant (five-fold) price-surge following the Brooklyn Subway Shooting on April 12 2022, in New York City. There was much controversy around the two ride-hailing firms’ algorithms’ decisions. The discussions evolved around various notions of fairness and the algorithms’ decisions’ justifiability. Our results indicate that algorithms, even if not explicitly addressed in the discourse, strongly impact on constructing fairness assessments and notions. They initiate the exchange, form people’s expectations, evoke people’s solidarity with specific groups, and are a vehicle for moral crusading. However, they are also subject to adjustments based on social forces. We claim that the process of constructing notions of fairness is no longer just social; it has become a socio-algorithmic process. We propose a theory of socio-algorithmic construction as a mechanism for establishing notions of fairness and other ethical constructs.

Workers’ occupational skill sets play a crucial role in successfully handling digital transformation. We investigate whether and how different types of occupational skill sets benefit from digital transformation. We theoretically and empirically analyze wage returns of workers in occupations with more or less specialized skill sets and with more or less social skills when IT increases in their industry. Applying natural language processing methods to the texts of occupational training curricula, we develop measures for occupational specialization and social skills. We use vocational education and training curricula from Switzerland because they cover approx. two-thirds of the working population. Using curricula, industry-level IT data and individual-level administrative wage data, our individual fixed-effects analyses show that IT progress leads to higher wage returns for workers in highly specialized occupations but not for workers in more general occupations. In addition, we find that high levels of social skills cannot make up for this difference when IT advances. However, our results indicate that for workers with high specialization, a combination with high social skills generates additional benefits when IT advances. Overall, our results suggest that, contrary to typical assumptions in educational policy debates, workers with specialized occupational skill sets - possibly in combination with high social skills - appear to be the ones who are particularly well prepared to cope with digital transformation.

We study whether outcome bias persists in markets with actors who are financially incentivized to make optimal decisions. We test whether inherently noisy match outcomes from European football are correctly incorporated into prices from a betting exchange market. We find that market prices overestimate (underestimate) the winning probability of teams that previously overperformed (underperformed) in terms of match outcomes compared to their performance based on “expected goals”. This pattern is mirrored in negative (positive) betting returns on overperforming (underperforming) teams. These results suggest that even competitive market mechanisms fail to completely erase outcome bias.

Why do residential mortgages carry a fixed or an adjustable interest rate? To answer this question we study unique data from 103 banks belonging to 73 different banking groups across twelve countries in the euro area. To explain the large cross-country and time variations observed, we distinguish between household conditions that determine the local demand for credit and the characteristics of banks that supply credit. As bank funding mostly occurs at the group level, we disentangle these two sets of factors by comparing the outcome observed for the same banking group across the different countries. Local household conditions dominate. In particular we find that the share of new loans with a fixed rate is larger when: (1) the historical volatility of inflation is lower, (2) the correlation between unemployment and the short-term interest rate is higher, (3) households’ financial literacy is lower, and (4) the use of local mortgages to back covered bonds and of mortgage-backed securities is more widespread.