In the evolving field of corporate sustainability, analyzing unstructured Environmental, Social, and Governance (ESG) reports is a complex challenge due to their varied formats and intricate content. This study introduces an innovative methodology utilizing the "Unstructured Core Library", specifically tailored to address these challenges by transforming ESG reports into structured, analyzable formats. Our approach significantly advances the existing research by offering high-precision text cleaning, adept identification and extraction of text from images, and standardization of tables within these reports. Emphasizing its capability to handle diverse data types, including text, images, and tables, the method adeptly manages the nuances of differing page layouts and report styles across industries. This research marks a substantial contribution to the fields of industrial ecology and corporate sustainability assessment, paving the way for the application of advanced NLP technologies and large language models in the analysis of corporate governance and sustainability. Our code is available at https://github.com/linancn/TianGong-AI-Unstructure.git.

In the ever-changing and dynamic realm of high-end fashion marketplaces, providing accurate and personalized size recommendations has become a critical aspect. Meeting customer expectations in this regard is not only crucial for ensuring their satisfaction but also plays a pivotal role in driving customer retention, which is a key metric for the success of any fashion retailer. We propose a novel sequence classification approach to address this problem, integrating implicit (Add2Bag) and explicit (ReturnReason) user signals. Our approach comprises two distinct models: one employs LSTMs to encode the user signals, while the other leverages an Attention mechanism. Our best model outperforms SFNet, improving accuracy by 45.7%. By using Add2Bag interactions we increase the user coverage by 24.5% when compared with only using Orders. Moreover, we evaluate the models' usability in real-time recommendation scenarios by conducting experiments to measure their latency performance.

As Large Language Models (LLMs) play an increasingly pivotal role in natural language processing applications, their safety concerns become critical areas of NLP research. This paper presents Safety and Over-Defensiveness Evaluation (SODE) benchmark: a collection of diverse safe and unsafe prompts with carefully designed evaluation methods that facilitate systematic evaluation, comparison, and analysis over 'safety' and 'over-defensiveness.' With SODE, we study a variety of LLM defense strategies over multiple state-of-the-art LLMs, which reveals several interesting and important findings, such as (a) the widely popular 'self-checking' techniques indeed improve the safety against unsafe inputs, but this comes at the cost of extreme over-defensiveness on the safe inputs, (b) providing a safety instruction along with in-context exemplars (of both safe and unsafe inputs) consistently improves safety and also mitigates undue over-defensiveness of the models, (c) providing contextual knowledge easily breaks the safety guardrails and makes the models more vulnerable to generating unsafe responses. Overall, our work reveals numerous such critical findings that we believe will pave the way and facilitate further research in improving the safety of LLMs.

This is the last weekly deal roundup we'll do this year, but unlike other end-of-year posts, we won't be looking back wistfully at the 2023 deals that once were -- all that matters are the ones you can get right now. This week, we're seeing a few sale prices that are still live from Black Friday; snag those before they inevitably go back up. A few new discounts have popped up that actually beat lows from November's shopping holiday, including a Prime-only deal on Amazon's Echo Show 8, all-time lows on Anker charging accessories, and discounts on Apple AirTags and Tile Pro trackers. Until next year, these are the best tech deals that you can still get today. As part of a larger Anker charging accessories sale, our top 30-watt fast charger is 30 percent off.

I was dreading buying Christmas gifts this year. My family tends to buy things they need as they go, and my sister would kill me if I bought her another sweater. So when my editor suggested I use ChatGPT to plan my Christmas shopping for me and write about it, I jumped at the opportunity. And I figured it was a win-win. If its suggested gifts were good, I wouldn't need to worry about coming up with present ideas for another 12 months! If they were a disaster, it would be a good opportunity to showcase how rudimentary artificial intelligence is (I'm extremely skeptical about the predictions of AI enslaving us in the future).

Large Language Models (LLMs) have demonstrated impressive potential to simulate human behavior. Using a causal inference framework, we empirically and theoretically analyze the challenges of conducting LLM-simulated experiments, and explore potential solutions. In the context of demand estimation, we show that variations in the treatment included in the prompt (e.g., price of focal product) can cause variations in unspecified confounding factors (e.g., price of competitors, historical prices, outside temperature), introducing endogeneity and yielding implausibly flat demand curves. We propose a theoretical framework suggesting this endogeneity issue generalizes to other contexts and won't be fully resolved by merely improving the training data. Unlike real experiments where researchers assign pre-existing units across conditions, LLMs simulate units based on the entire prompt, which includes the description of the treatment. Therefore, due to associations in the training data, the characteristics of individuals and environments simulated by the LLM can be affected by the treatment assignment. We explore two potential solutions. The first specifies all contextual variables that affect both treatment and outcome, which we demonstrate to be challenging for a general-purpose LLM. The second explicitly specifies the source of treatment variation in the prompt given to the LLM (e.g., by informing the LLM that the store is running an experiment). While this approach only allows the estimation of a conditional average treatment effect that depends on the specific experimental design, it provides valuable directional results for exploratory analysis.

This paper proposes a novel method for demand forecasting in a pricing context. Here, modeling the causal relationship between price as an input variable to demand is crucial because retailers aim to set prices in a (profit) optimal manner in a downstream decision making problem. Our methods bring together the Double Machine Learning methodology for causal inference and state-of-the-art transformer-based forecasting models. In extensive empirical experiments, we show on the one hand that our method estimates the causal effect better in a fully controlled setting via synthetic, yet realistic data. On the other hand, we demonstrate on real-world data that our method outperforms forecasting methods in off-policy settings (i.e., when there's a change in the pricing policy) while only slightly trailing in the on-policy setting.

While it's a bit too late to receive most gifts in time for Christmas, there are still a handful of good gadget deals floating around if you're shopping for yourself. If you need a new laptop today, for instance, multiple configurations of the 13-inch MacBook Air are $200 off Apple's list price. The 15-inch Air, meanwhile, is available for as low as $999, a $300 discount. A bundle of Apple's AirTags is down to $79, while a pack of Tile trackers is down to $50. The Xbox Series X is still $150 off, and the major video game storefronts have kicked off their annual winter sales, with sweeping discounts across Steam, the Nintendo eShop, the PlayStation Store and the Microsoft Store.

Significant research effort has been devoted in recent years to developing personalized pricing, promotions, and product recommendation algorithms that can leverage rich customer data to learn and earn. Systematic benchmarking and evaluation of these causal learning systems remains a critical challenge, due to the lack of suitable datasets and simulation environments. In this work, we propose a multi-stage model for simulating customer shopping behavior that captures important sources of heterogeneity, including price sensitivity and past experiences. We embedded this model into a working simulation environment -- RetailSynth. RetailSynth was carefully calibrated on publicly available grocery data to create realistic synthetic shopping transactions. Multiple pricing policies were implemented within the simulator and analyzed for impact on revenue, category penetration, and customer retention. Applied researchers can use RetailSynth to validate causal demand models for multi-category retail and to incorporate realistic price sensitivity into emerging benchmarking suites for personalized pricing, promotions, and product recommendations.

The extraction of a small number of relevant insights from vast amounts of data is a crucial component of data-driven decision-making. However, accomplishing this task requires considerable technical skills, domain expertise, and human labor. This study explores the potential of using Large Language Models (LLMs) to automate the discovery of insights in data, leveraging recent advances in reasoning and code generation techniques. We propose a new evaluation methodology based on a "capture the flag" principle, measuring the ability of such models to recognize meaningful and pertinent information (flags) in a dataset. We further propose two proof-of-concept agents, with different inner workings, and compare their ability to capture such flags in a real-world sales dataset. While the work reported here is preliminary, our results are sufficiently interesting to mandate future exploration by the community.