ABSTRACT Large language models (LLMs) have shown strong reasoning capabilities and are increasingly explored for financial trading. Existing LLM-based trading agents, however, largely focus on single-step prediction and lack integrated mechanisms for risk management, which reduces their effectiveness in volatile markets. We introduce FinRS, a risk-sensitive trading framework that combines hierarchical market analysis, dual-decision agents, and multi-timescale reward reflection to align trading actions with both return objectives and downside risk constraints. Experiments on multiple stocks and market conditions show that FinRS achieves superior profitability and stability compared to state-of-the-art methods. Index T erms-- Trading Agent, Risk-Sensitive, Real Markets 1. INTRODUCTION In recent years, large language models (LLMs) [1, 2] have demonstrated significant potential in financial trading.

Purpose: This study introduces a novel framework for identifying and exploiting predictive lead-lag relationships in financial markets. We propose an integrated approach that combines advanced statistical methodologies with machine learning models to enhance the identification and exploitation of predictive relationships between equities. Methods: We employed a Gaussian Mixture Model (GMM) to cluster nine prominent stocks based on their mid-range historical volatility profiles over a three-year period. From the resulting clusters, we constructed a multi-stage causal inference pipeline, incorporating the Granger Causality Test (GCT), a customised Peter-Clark Momentary Conditional Independence (PCMCI) test, and Effective Transfer Entropy (ETE) to identify robust, predictive linkages. Subsequently, Dynamic Time Warping (DTW) and a K-Nearest Neighbours (KNN) classifier were utilised to determine the optimal time lag for trade execution. The resulting strategy was rigorously backtested. Results: The proposed volatility-based trading strategy, tested from 8 June 2023 to 12 August 2023, demonstrated substantial efficacy. The portfolio yielded a total return of 15.38%, significantly outperforming the 10.39% return of a comparative Buy-and-Hold strategy. Key performance metrics, including a Sharpe Ratio up to 2.17 and a win rate up to 100% for certain pairs, confirmed the strategy's viability. Conclusion: This research contributes a systematic and robust methodology for identifying profitable trading opportunities derived from volatility-based causal relationships. The findings have significant implications for both academic research in financial modelling and the practical application of algorithmic trading, offering a structured approach to developing resilient, data-driven strategies.

Recent advancements have underscored the potential of large language model (LLM)-based agents in financial decision-making. Despite this progress, the field currently encounters two main challenges: (1) the lack of a comprehensive LLM agent framework adaptable to a variety of financial tasks, and (2) the absence of standardized benchmarks and consistent datasets for assessing agent performance. To tackle these issues, we introduce \textsc{InvestorBench}, the first benchmark specifically designed for evaluating LLM-based agents in diverse financial decision-making contexts. InvestorBench enhances the versatility of LLM-enabled agents by providing a comprehensive suite of tasks applicable to different financial products, including single equities like stocks, cryptocurrencies and exchange-traded funds (ETFs). Additionally, we assess the reasoning and decision-making capabilities of our agent framework using thirteen different LLMs as backbone models, across various market environments and tasks. Furthermore, we have curated a diverse collection of open-source, multi-modal datasets and developed a comprehensive suite of environments for financial decision-making. This establishes a highly accessible platform for evaluating financial agents' performance across various scenarios.

Tesla (TSLA) will hold its second annual AI Day in Palo Alto, California, Friday evening. The six-hour event will include updates on Tesla (TSLA)'s work in artificial intelligence, "Full Self-Driving," its supercomputer "Dojo" and maybe a humanoid robot, according to invitations posted online by Tesla (TSLA) supporters. The event is expected to be live-streamed. Dojo is a supercomputer being designed to train AI systems to complete complex tasks like Tesla's driver-assistance systems Autopilot and "Full Self-Driving," which sometimes perform some driving tasks like steering and keeping up with traffic. Tesla's previous AI Day included detailed technical explanations of the company's work in a bid to attract leading engineers.

Welcome to Part 3 of the series. Part 1 covers Tesla's manufacturing advantage. Part 2 covers Tesla's energy business. In this article, I will give my assessment of Tesla's Artificial Intelligence efforts and my take on its bottom line. These are just my opinions.

Label smoothing regularization (LSR) has a great success in training deep neural networks by stochastic algorithms such as stochastic gradient descent and its variants. However, the theoretical understanding of its power from the view of optimization is still rare. This study opens the door to a deep understanding of LSR by initiating the analysis. In this paper, we analyze the convergence behaviors of stochastic gradient descent with label smoothing regularization for solving non-convex problems and show that an appropriate LSR can help to speed up the convergence by reducing the variance. More interestingly, we proposed a simple yet effective strategy, namely Two-Stage LAbel smoothing algorithm (TSLA), that uses LSR in the early training epochs and drops it off in the later training epochs. We observe from the improved convergence result of TSLA that it benefits from LSR in the first stage and essentially converges faster in the second stage. To the best of our knowledge, this is the first work for understanding the power of LSR via establishing convergence complexity of stochastic methods with LSR in non-convex optimization. We empirically demonstrate the effectiveness of the proposed method in comparison with baselines on training ResNet models over benchmark data sets.

It's usually a mistake to bet against Elon Musk. The recent performance of Tesla (NASDAQ:TSLA) stock is a good example. True, Elon Musk may say some wacky things and make some big mistakes. But in the end, he always seems to find ways to achieve his lofty goals. Earlier in the year, TSLA appeared to be in a bleak situation, and there was many questions about its outlook.

The US dollar had an event-heavy week to start off June. The US dollar surged early last week as the uncertainty about the Euro arose due to political events happened in Europe and the volatility in Asian markets driven by threats of an immediate trade war between the US and China. On Thursday (May 31), the Euro rebounded as Italy's politicians seemed to have found a resolution to their struggles in forming a new government. In the same day, the Trump administration announced it was putting tariffs on steel and aluminum imports from Canada, Mexico and Europe, strengthening fears over the trade war and making the US dollar suffer a slump. The US labor indicators highlighted the fundamental strength of the country's economy and made the US dollar extend gains amid the Europe geopolitical turmoil.

Two weeks ago, Tesla Motors CEO Elon Musk teased on Twitter an upcoming "P100D Ludicrous Easter egg" update that would unlock the full performance of the Model S. And on Wednesday, he announced exactly what that Easter egg is. When the P100D was announced in August, it already broke records, not just for electric vehicles but for car speeds in general. It could accelerate from 0 to 60 miles an hour in 2.5 seconds, making it the fastest car in the world (excluding supercars). This update shaves an extra 0.1 second off that time.

Analysts expect Tesla Motors (TSLA) to show that it has reigned in its losses when the automaker releases its third quarter earnings Wednesday after more than doubling analysts' loss expectations last quarter. Investors polled by FactSet expected the California-based company, led by South African investor and engineer Elon Musk, to post a loss of 53 cents per share--an improvement from last year's third quarter losses of 1.78 per share and second quarter losses of 1.06 per share. Analysts surveyed by Reuters, however, estimated the company's earnings per share (EPS) would turn positive, hitting 37 cents for the quarter ending in December. Such optimistic forecasts are likely fueled by Tesla's announcement that it saw a 70 percent rise in third-quarter deliveries early in October--welcome news after, two months earlier, it disclosed third quarter expenditure needs of 1.1 billion. The earnings report will come one week after the automaker announced all of its cars currently in production would be self-driving, with "a safety level substantially greater than that of a human driver."