Tracking the strategic focus of companies through topics in their earnings calls is a key task in financial analysis. However, as industries evolve, traditional topic modeling techniques struggle to dynamically capture emerging topics and their relationships. In this work, we propose an LLM-agent driven approach to discover and retrieve emerging topics from quarterly earnings calls. We propose an LLM-agent to extract topics from documents, structure them into a hierarchical ontology, and establish relationships between new and existing topics through a topic ontology. We demonstrate the use of extracted topics to infer company-level insights and emerging trends over time. We evaluate our approach by measuring ontology coherence, topic evolution accuracy, and its ability to surface emerging financial trends.

Nvidia has notched a market capitalisation of 4 trillion, making it the first public company in the world to reach the milestone and solidifying its position as one of Wall Street's most-favoured stocks. On Wednesday, shares of the leading chip designer rose as much as 2.5 percent to an all-time high of 164, benefiting from the continuing surge in demand for artificial intelligence technologies. The stock's recent rally comes despite a sluggish start to the year, when the emergence of a Chinese discount artificial intelligence model developed by DeepSeek shook confidence in stocks linked to the sector. Nvidia achieved a 1 trillion market value for the first time in June 2023 and tripled it in about a year, faster than Apple and Microsoft, the only other United States firms with a market value of more than 3 trillion. Microsoft is the second-biggest US company, with a market capitalisation of 3.75 trillion.

Extracting structured and quantitative insights from unstructured financial filings is essential in investment research, yet remains time-consuming and resource-intensive. Conventional approaches in practice rely heavily on labor-intensive manual processes, limiting scalability and delaying the research workflow. In this paper, we propose an efficient and scalable method for accurately extracting quantitative insights from unstructured financial documents, leveraging a multi-agent system composed of large language models. Our proposed multi-agent system consists of two specialized agents: the \emph{Extraction Agent} and the \emph{Text-to-SQL Agent}. The \textit{Extraction Agent} automatically identifies key performance indicators from unstructured financial text, standardizes their formats, and verifies their accuracy. On the other hand, the \textit{Text-to-SQL Agent} generates executable SQL statements from natural language queries, allowing users to access structured data accurately without requiring familiarity with the database schema. Through experiments, we demonstrate that our proposed system effectively transforms unstructured text into structured data accurately and enables precise retrieval of key information. First, we demonstrate that our system achieves approximately 95\% accuracy in transforming financial filings into structured data, matching the performance level typically attained by human annotators. Second, in a human evaluation of the retrieval task -- where natural language queries are used to search information from structured data -- 91\% of the responses were rated as correct by human evaluators. In both evaluations, our system generalizes well across financial document types, consistently delivering reliable performance.

The video game industry loves to tout figures: record-breaking sales numbers, astonishing revenue growth, dazzling quantities of concurrent players. It makes sense that the people who make and play games love numbers: They're proof that someone is winning. We have a new incredible number from the world of video games: In spite of an alarming price tag, it took only four days for the Nintendo Switch 2 to become the fastest-selling home video game console of all time, with 3.5 million units sold over the weekend following its June 5 release. This is tremendous business, enough for investors to take note and consider Nintendo a safe haven in a moment of extreme economic volatility. This kind of success is typically a point of pride to proponents of the video game industry, hard data proving the medium's significance to any doubters.

Law has long been a domain that has been popular in natural language processing (NLP) applications. Reasoning (ratiocination and the ability to make connections to precedent) is a core part of the practice of the law in the real world. Nevertheless, while multiple legal datasets exist, none have thus far focused specifically on reasoning tasks. We focus on a specific aspect of the legal landscape by introducing a corporate governance reasoning benchmark (CHANCERY) to test a model's ability to reason about whether executive/board/shareholder's proposed actions are consistent with corporate governance charters. This benchmark introduces a first-of-its-kind corporate governance reasoning test for language models - modeled after real world corporate governance law. The benchmark consists of a corporate charter (a set of governing covenants) and a proposal for executive action. The model's task is one of binary classification: reason about whether the action is consistent with the rules contained within the charter. We create the benchmark following established principles of corporate governance - 24 concrete corporate governance principles established in and 79 real life corporate charters selected to represent diverse industries from a total dataset of 10k real life corporate charters. Evaluations on state-of-the-art (SOTA) reasoning models confirm the difficulty of the benchmark, with models such as Claude 3.7 Sonnet and GPT-4o achieving 64.5% and 75.2% accuracy respectively. Reasoning agents exhibit superior performance, with agents based on the ReAct and CodeAct frameworks scoring 76.1% and 78.1% respectively, further confirming the advanced legal reasoning capabilities required to score highly on the benchmark. We also conduct an analysis of the types of questions which current reasoning models struggle on, revealing insights into the legal reasoning capabilities of SOTA models.

Taiwan's Yageo said it would keep Shibaura Electronics's most advanced technology in Japan if it successfully acquires the artificial intelligence sensor maker. The comments from Yageo founder and Chairman Pierre Chen come as Tokyo seeks to strike a balance between shareholder returns while ensuring cutting-edge AI technology stays at home. Shibaura's high-precision thermistors are key for monitoring the internal temperature of electronic devices to prevent overheating. That's especially important in AI, where data centers with large clusters of high-performance servers churn through troves of data. "It is not in Yageo's interest to see Shibaura's technology transfer to countries that Japan considers to be unfriendly," Chen told reporters in Taipei on Saturday.

Nvidia beat Wall Street expectations in its quarterly earnings report on Wednesday, marking another in a string of financial wins for the computer hardware giant. It reported 44.1bn in revenue in the quarter ending in April, up 69% from the previous year. The company exceeded investors' predictions of 43.3bn in revenue. Adjusted earnings per share came in at 0.81, under investor expectations of an adjusted earnings per share of 88 cents. The company also reported 39.1bn in data center revenue, up 73% from the year prior.

Recent large reasoning models (LRMs) have demonstrated strong reasoning capabilities through reinforcement learning (RL). These improvements have primarily been observed within the short-context reasoning tasks. In contrast, extending LRMs to effectively process and reason on long-context inputs via RL remains a critical unsolved challenge. To bridge this gap, we first formalize the paradigm of long-context reasoning RL, and identify key challenges in suboptimal training efficiency and unstable optimization process. To address these issues, we propose QwenLong-L1, a framework that adapts short-context LRMs to long-context scenarios via progressive context scaling. Specifically, we utilize a warm-up supervised fine-tuning (SFT) stage to establish a robust initial policy, followed by a curriculum-guided phased RL technique to stabilize the policy evolution, and enhanced with a difficulty-aware retrospective sampling strategy to incentivize the policy exploration. Experiments on seven long-context document question-answering benchmarks demonstrate that QwenLong-L1-32B outperforms flagship LRMs like OpenAI-o3-mini and Qwen3-235B-A22B, achieving performance on par with Claude-3.7-Sonnet-Thinking, demonstrating leading performance among state-of-the-art LRMs. This work advances the development of practical long-context LRMs capable of robust reasoning across information-intensive environments.

Fact-checking is extensively studied in the context of misinformation and disinformation, addressing objective inaccuracies. However, a softer form of misinformation involves responses that are factually correct but lack certain features such as clarity and relevance. This challenge is prevalent in formal Question-Answer (QA) settings such as press conferences in finance, politics, sports, and other domains, where subjective answers can obscure transparency. Despite this, there is a lack of manually annotated datasets for subjective features across multiple dimensions. To address this gap, we introduce SubjECTive-QA, a human annotated dataset on Earnings Call Transcripts' (ECTs) QA sessions as the answers given by company representatives are often open to subjective interpretations and scrutiny.

People in the real world often possess vague knowledge of future payoffs, for which quantification is not feasible or desirable. We argue that language, with differing ability to convey vague information, plays an important but less-known role in representing subjective expectations. Empirically, we find that in their reports, analysts include useful information in linguistic expressions but not numerical forecasts. Specifically, the textual tone of analyst reports has predictive power for forecast errors and subsequent revisions in numerical forecasts, and this relation becomes stronger when analyst's language is vaguer, when uncertainty is higher, and when analysts are busier. Overall, our theory and evidence suggest that some useful information is vaguely known and only communicated through language.