Pre-trained language models (PLMs) are trained on large amounts of data, which helps capture world knowledge alongside linguistic competence. Due to this, they are extensively used for ultra-fine entity typing tasks, where they provide the entity knowledge held in its parameter space. Given that PLMs learn from co-occurrence patterns, they likely contain more knowledge or less knowledge about entities depending on their how frequent they are in the pre-training data. In this work, we probe PLMs to elicit encoded entity probabilities and demonstrate that they highly correlate with their frequency in large-scale internet data. Then, we demonstrate that entity-typing approaches that rely on PLMs struggle with entities at the long tail on the distribution. Our findings suggests that we need to go beyond PLMs to produce solutions that perform well for rare, new or infrequent entities.

This paper introduces the UCFE: User-Centric Financial Expertise benchmark, an innovative framework designed to evaluate the ability of large language models (LLMs) to handle complex real-world financial tasks. UCFE benchmark adopts a hybrid approach that combines human expert evaluations with dynamic, task-specific interactions to simulate the complexities of evolving financial scenarios. Firstly, we conducted a user study involving 804 participants, collecting their feedback on financial tasks. Secondly, based on this feedback, we created our dataset that encompasses a wide range of user intents and interactions. This dataset serves as the foundation for benchmarking 12 LLM services using the LLM-as-Judge methodology. Our results show a significant alignment between benchmark scores and human preferences, with a Pearson correlation coefficient of 0.78, confirming the effectiveness of the UCFE dataset and our evaluation approach. UCFE benchmark not only reveals the potential of LLMs in the financial sector but also provides a robust framework for assessing their performance and user satisfaction. The benchmark dataset and evaluation code are available.

Enabling Large Language Models (LLMs) to handle a wider range of complex tasks (e.g., coding, math) has drawn great attention from many researchers. As LLMs continue to evolve, merely increasing the number of model parameters yields diminishing performance improvements and heavy computational costs. Recently, OpenAI's o1 model has shown that inference strategies (i.e., Test-time Compute methods) can also significantly enhance the reasoning capabilities of LLMs. However, the mechanisms behind these methods are still unexplored. In our work, to investigate the reasoning patterns of o1, we compare o1 with existing Test-time Compute methods (BoN, Step-wise BoN, Agent Workflow, and Self-Refine) by using OpenAI's GPT-4o as a backbone on general reasoning benchmarks in three domains (i.e., math, coding, commonsense reasoning). Specifically, first, our experiments show that the o1 model has achieved the best performance on most datasets. Second, as for the methods of searching diverse responses (e.g., BoN), we find the reward models' capability and the search space both limit the upper boundary of these methods. Third, as for the methods that break the problem into many sub-problems, the Agent Workflow has achieved better performance than Step-wise BoN due to the domain-specific system prompt for planning better reasoning processes. Fourth, it is worth mentioning that we have summarized six reasoning patterns of o1, and provided a detailed analysis on several reasoning benchmarks.

Recent advancements in tool-equipped Agents (LLMs) have enabled complex tasks like secure database interactions and multi-agent code development. However, scaling tool capacity beyond agent reasoning or model limits remains a challenge. In this paper, we address these challenges by introducing Toolshed Knowledge Bases, a tool knowledge base (vector database) designed to store enhanced tool representations and optimize tool selection for large-scale tool-equipped Agents. Additionally, we propose Advanced RAG-Tool Fusion, a novel ensemble of tool-applied advanced retrieval-augmented generation (RAG) techniques across the pre-retrieval, intra-retrieval, and post-retrieval phases, without requiring model fine-tuning. During pre-retrieval, tool documents are enhanced with key information and stored in the Toolshed Knowledge Base. Intra-retrieval focuses on query planning and transformation to increase retrieval accuracy. Post-retrieval refines the retrieved tool documents and enables self-reflection. Furthermore, by varying both the total number of tools (tool-M) an Agent has access to and the tool selection threshold (top-k), we address trade-offs between retrieval accuracy, agent performance, and token cost. Our approach achieves 46%, 56%, and 47% absolute improvements on the ToolE single-tool, ToolE multi-tool and Seal-Tools benchmark datasets, respectively (Recall@5).

Proximal Policy Optimization (PPO) is commonly used in Reinforcement Learning from Human Feedback to align large language models (LLMs) with downstream tasks. This paper investigates the feasibility of using PPO for direct reinforcement learning (RL) from explicitly programmed reward signals, as opposed to indirect learning from human feedback via an intermediary reward model. We focus on tasks expressed through formal languages, such as mathematics and programming, where explicit reward functions can be programmed to automatically assess the quality of generated outputs. We apply this approach to a sentiment alignment task, a simple arithmetic task, and a more complex game synthesis task. The sentiment alignment task replicates prior research and serves to validate our experimental setup. Our results show that pure RL-based training for the two formal language tasks is challenging, with success being limited even for the simple arithmetic task. We propose a novel batch-entropy regularization term to aid exploration, although training is not yet entirely stable. Our findings suggest that direct RL training of LLMs may be more suitable for relatively minor changes, such as alignment, than for learning new tasks altogether, even if an informative reward signal can be expressed programmatically.

Enabling robots to autonomously navigate unknown, complex, dynamic environments and perform diverse tasks remains a fundamental challenge in developing robust autonomous physical agents. These agents must effectively perceive their surroundings while leveraging world knowledge for decision-making. Although recent approaches utilize vision-language and large language models for scene understanding and planning, they often rely on offline processing, offboard compute, make simplifying assumptions about the environment and perception, limiting real-world applicability. We present a novel framework for real-time onboard autonomous navigation in unknown environments that change over time by integrating multi-level abstraction in both perception and planning pipelines. Our system fuses data from multiple onboard sensors for localization and mapping and integrates it with open-vocabulary semantics to generate hierarchical scene graphs from continuously updated semantic object map. The LLM-based planner uses these graphs to create multi-step plans that guide low-level controllers in executing navigation tasks specified in natural language. The system's real-time operation enables the LLM to adjust its plans based on updates to the scene graph and task execution status, ensuring continuous adaptation to new situations or when the current plan cannot accomplish the task, a key advantage over static or rule-based systems. We demonstrate our system's efficacy on a quadruped navigating dynamic environments, showcasing its adaptability and robustness in diverse scenarios.

Multilingual Large Language Models (MLLMs) represent a pivotal advancement in democratizing artificial intelligence across linguistic boundaries. While theoretical foundations are well-established, practical implementation guidelines remain scattered. This work bridges this gap by providing a comprehensive end-to-end framework for developing and deploying MLLMs in production environments. We make three distinctive contributions: First, we present an actionable pipeline from data pre-processing through deployment, integrating insights from academic research and industrial applications. Second, using Llama2 as a case study, we provide detailed optimization strategies for enhancing multilingual capabilities, including curriculum learning approaches for balancing high-resource and low-resource languages, tokenization strategies, and effective sampling methods. Third, we offer an interdisciplinary analysis that considers technical, linguistic, and cultural perspectives in MLLM development. Our findings reveal critical challenges in supporting linguistic diversity, with 88.38% of world languages categorized as low-resource, affecting over a billion speakers. We examine practical solutions through real-world applications in customer service, search engines, and machine translation. By synthesizing theoretical frameworks with production-ready implementation strategies, this survey provides essential guidance for practitioners and researchers working to develop more inclusive and effective multilingual AI systems.

Explainable Artificial Intelligence (AI) focuses on helping humans understand the working of AI systems or their decisions and has been a cornerstone of AI for decades. Recent research in explainability has focused on explaining the workings of AI models or model explainability. There have also been several position statements and review papers detailing the needs of end-users for user-centered explainability but fewer implementations. Hence, this thesis seeks to bridge some gaps between model and user-centered explainability. We create an explanation ontology (EO) to represent literature-derived explanation types via their supporting components. We implement a knowledge-augmented question-answering (QA) pipeline to support contextual explanations in a clinical setting. Finally, we are implementing a system to combine explanations from different AI methods and data modalities. Within the EO, we can represent fifteen different explanation types, and we have tested these representations in six exemplar use cases. We find that knowledge augmentations improve the performance of base large language models in the contextualized QA, and the performance is variable across disease groups. In the same setting, clinicians also indicated that they prefer to see actionability as one of the main foci in explanations. In our explanations combination method, we plan to use similarity metrics to determine the similarity of explanations in a chronic disease detection setting. Overall, through this thesis, we design methods that can support knowledge-enabled explanations across different use cases, accounting for the methods in today's AI era that can generate the supporting components of these explanations and domain knowledge sources that can enhance them.

Attacking fairness is crucial because compromised models can introduce biased outcomes, undermining trust and amplifying inequalities in sensitive applications like hiring, healthcare, and law enforcement. This highlights the urgent need to understand how fairness mechanisms can be exploited and to develop defenses that ensure both fairness and robustness. We introduce BadFair, a novel backdoored fairness attack methodology. BadFair stealthily crafts a model that operates with accuracy and fairness under regular conditions but, when activated by certain triggers, discriminates and produces incorrect results for specific groups. This type of attack is particularly stealthy and dangerous, as it circumvents existing fairness detection methods, maintaining an appearance of fairness in normal use. Our findings reveal that BadFair achieves a more than 85% attack success rate in attacks aimed at target groups on average while only incurring a minimal accuracy loss. Moreover, it consistently exhibits a significant discrimination score, distinguishing between pre-defined target and non-target attacked groups across various datasets and models.

The increasing popularity of Large Language Models (LLMs) in recent years has changed the way users interact with and pose questions to AI-based conversational systems. An essential aspect for increasing the trustworthiness of generated LLM answers is the ability to trace the individual claims from responses back to relevant sources that support them, the process known as answer attribution. While recent work has started exploring the task of answer attribution in LLMs, some challenges still remain. In this work, we first perform a case study analyzing the effectiveness of existing answer attribution methods, with a focus on subtasks of answer segmentation and evidence retrieval. Based on the observed shortcomings, we propose new methods for producing more independent and contextualized claims for better retrieval and attribution. The new methods are evaluated and shown to improve the performance of answer attribution components. We end with a discussion and outline of future directions for the task.