Understanding sentiment in financial documents is crucial for gaining insights into market behavior. These reports often contain obfuscated language designed to present a positive or neutral outlook, even when underlying conditions may be less favorable. This paper presents a novel approach using Aspect-Based Sentiment Analysis (ABSA) to decode obfuscated sentiment in Thai financial annual reports. We develop specific guidelines for annotating obfuscated sentiment in these texts and annotate more than one hundred financial reports. We then benchmark various text classification models on this annotated dataset, demonstrating strong performance in sentiment classification. Additionally, we conduct an event study to evaluate the real-world implications of our sentiment analysis on stock prices. Our results suggest that market reactions are selectively influenced by specific aspects within the reports. Our findings underscore the complexity of sentiment analysis in financial texts and highlight the importance of addressing obfuscated language to accurately assess market sentiment.

--Corporate credit rating serves as a crucial intermediary service in the market economy, playing a key role in maintaining economic order . Existing credit rating models rely on financial metrics and deep learning. However, they often overlook insights from non-financial data, such as corporate annual reports. T o address this, this paper introduces a corporate credit rating framework that integrates financial data with features extracted from annual reports using FinBERT, aiming to fully leverage the potential value of unstructured text data. In addition, we have developed a large-scale dataset, the Comprehensive Corporate Rating Dataset (CCRD), which combines both traditional financial data and textual data from annual reports. The experimental results show that the proposed method improves the accuracy of the rating predictions by 8-12%, significantly improving the effectiveness and reliability of corporate credit ratings.

Financial analysis presents complex challenges that could leverage large language model (LLM) capabilities. However, the scarcity of challenging financial datasets, particularly for Japanese financial data, impedes academic innovation in financial analytics. As LLMs advance, this lack of accessible research resources increasingly hinders their development and evaluation in this specialized domain. To address this gap, we introduce EDINET-Bench, an open-source Japanese financial benchmark designed to evaluate the performance of LLMs on challenging financial tasks including accounting fraud detection, earnings forecasting, and industry prediction. EDINET-Bench is constructed by downloading annual reports from the past 10 years from Japan's Electronic Disclosure for Investors' NETwork (EDINET) and automatically assigning labels corresponding to each evaluation task. Our experiments reveal that even state-of-the-art LLMs struggle, performing only slightly better than logistic regression in binary classification for fraud detection and earnings forecasting. These results highlight significant challenges in applying LLMs to real-world financial applications and underscore the need for domain-specific adaptation. Our dataset, benchmark construction code, and evaluation code is publicly available to facilitate future research in finance with LLMs.

Recent advances in Large Language Models (LLMs) have enabled them to process increasingly longer sequences, ranging from 2K to 2M tokens and even beyond. However, simply extending the input sequence length does not necessarily lead to effective long-context understanding. In this study, we integrate Chain-of-Thought (CoT) reasoning into LLMs in a supervised manner to facilitate effective long-context understanding. To achieve this, we introduce LongFinanceQA, a synthetic dataset in the financial domain designed to improve long-context reasoning. Unlike existing long-context synthetic data, LongFinanceQA includes intermediate CoT reasoning before the final conclusion, which encourages LLMs to perform explicit reasoning, improving accuracy and interpretability in long-context understanding. To generate synthetic CoT reasoning, we propose Property-driven Agentic Inference (PAI), an agentic framework that simulates human-like reasoning steps, including property extraction, retrieval, and summarization. We evaluate PAI's reasoning capabilities by assessing GPT-4o-mini w/ PAI on the Loong benchmark, outperforming standard GPT-4o-mini by 20.0%. Furthermore, we fine-tune LLaMA-3.1-8B-Instruct on LongFinanceQA, achieving a 24.6% gain on Loong's financial subset.

The H-1B visa program was intended to bring in specialized talent from abroad, but instead it has become a tool for employers to hire lower-cost labor for ordinary jobs. The result is a distorted job market, where highly skilled workers are being squeezed out of the H-1B visa program by spam applications for ordinary workers who then take entry-level positions that are already in short supply. This misuse of H-1B visas has a negative synergy with growing impact of AI on the job market and is part of a larger problem that urgently needs attention. The impact of this visa-farming problem is particularly acute among young people and recent college graduates, who face a bleak job market despite moderate overall unemployment rates. According to government data, the ratio of unemployment for college grads under 25 to those over 25 has hit an all-time high of more than four to one.

For assessing various performance indicators of companies, the focus is shifting from strictly financial (quantitative) publicly disclosed information to qualitative (textual) information. This textual data can provide valuable weak signals, for example through stylistic features, which can complement the quantitative data on financial performance or on Environmental, Social and Governance (ESG) criteria. In this work, we use various multi-task learning methods for financial text classification with the focus on financial sentiment, objectivity, forward-looking sentence prediction and ESG-content detection. We propose different methods to combine the information extracted from training jointly on different tasks; our best-performing method highlights the positive effect of explicitly adding auxiliary task predictions as features for the final target task during the multi-task training. Next, we use these classifiers to extract textual features from annual reports of FTSE350 companies and investigate the link between ESG quantitative scores and these features.

Financial regulators in the United States have named artificial intelligence (AI) as a risk to the financial system for the first time. In its latest annual report, the Financial Stability Oversight Council said the growing use of AI in financial services is a "vulnerability" that should be monitored. While AI offers the promise of reducing costs, improving efficiency, identifying more complex relationships and improving performance and accuracy, it can also "introduce certain risks, including safety-and-soundness risks like cyber and model risks," the FSOC said in its annual report released on Thursday. The FSOC, which was established in the wake of the 2008 financial crisis to identify excessive risks in the financial system, said developments in AI should be monitored to ensure that oversight mechanisms "account for emerging risks" while facilitating "efficiency and innovation". Authorities must also "deepen expertise and capacity" to monitor the field, the FSOC said.

In this work, we examine the evaluation process for the task of detecting financial reports with a high risk of containing a misstatement. This task is often referred to, in the literature, as ``misstatement detection in financial reports''. We provide an extensive review of the related literature. We propose a new, realistic evaluation framework for the task which, unlike a large part of the previous work: (a) focuses on the misstatement class and its rarity, (b) considers the dimension of time when splitting data into training and test and (c) considers the fact that misstatements can take a long time to detect. Most importantly, we show that the evaluation process significantly affects system performance, and we analyze the performance of different models and feature types in the new realistic framework.

Hedges are widely studied across registers and disciplines, yet research on the translation of hedges in political texts is extremely limited. This contrastive study is dedicated to investigating whether there is a diachronic change in the frequencies of hedging devices in the target texts, to what extent the changing frequencies of translated hedges through years are attributed to the source texts, and what translation strategies are adopted to deal with them. For the purposes of this research, two types of official political texts and their translations from China and the United Nations were collected to form three sub-corpora. Results show that hedges tend to appear more frequently in English political texts, be it original English or translated English. In addition, directionality seems to play an important role in influencing both the frequencies and translation strategies regarding the use of hedges. A noticeable diachronic increase of hedging devices is also observed in our corpus.

Regarding business outcomes, the company claims that a large international information-services firm reduced the time spent on legal reviews by 75%. In another example, DocuSign cited how they decreased the time an international telecom company spent reviewing customer agreements by more than 80%, and enabled a global financial services leader to automate the analysis of over 2.6 million data points from supplier agreements. It's telling that the company can only procure a handful of examples and not one is willing to be named. Resolutely successful initiatives usually have no problem finding a dozen companies willing to lend their name – even to a banner on a company front page – to a brand that authentically benefited them. It's also worth noting that DocuSign's Rolodex is hardly wanting: the company lists T-Mobile, Unilever, Boston Scientific, AAA, and Salesforce as some of their past clients.