Knowledge comprehension capability is an important aspect of human intelligence. As Large Language Models (LLMs) are being envisioned as superhuman agents, it is crucial for them to be proficient at knowledge comprehension. However, existing benchmarking studies do not provide consistent, generalizable, and formal guarantees on the knowledge comprehension capabilities of LLMs. In this work, we propose the first framework to certify knowledge comprehension in LLMs with formal probabilistic guarantees. Our certificates are quantitative -- they consist of high-confidence, tight bounds on the probability that a target LLM gives the correct answer on any knowledge comprehension prompt sampled from a distribution. We design and certify novel specifications that precisely represent distributions of knowledge comprehension prompts leveraging knowledge graphs. We certify SOTA LLMs for specifications over the Wikidata5m knowledge graph. We find that the knowledge comprehension capability improves significantly with scaling the size of the models.
相關內容
通過學習(xi)、實踐或探索(suo)所獲得(de)的認識(shi)、判斷或技(ji)能(neng)。
Quantifying and Mitigating Unimodal Biases in Multimodal Large Language Models: A Causal Perspective
Recent advancements in Large Language Models (LLMs) have facilitated the development of Multimodal LLMs (MLLMs). Despite their impressive capabilities, MLLMs often suffer from over-reliance on unimodal biases (e.g., language bias and vision bias), leading to incorrect answers or hallucinations in complex multimodal tasks. To investigate this issue, we propose a causal framework to interpret the biases in Visual Question Answering (VQA) problems. Within this framework, we conduct an in-depth causal analysis to assess the causal effect of these biases on MLLM predictions. Based on the analysis, we introduce 1) a novel MORE dataset with 12,000 challenging VQA instances requiring multi-hop reasoning and overcoming unimodal biases. 2) a causality-enhanced agent framework CAVE that guides models to comprehensively integrate information from different modalities and mitigate biases. Our experiments show that MLLMs perform poorly on MORE, indicating strong unimodal biases and limited semantic understanding. However, when integrated with our CAVE, promising improvements in reasoning and bias mitigation can be seen. These findings provide important insights for the development of more robust MLLMs and contribute to the broader goal of advancing multimodal AI systems capable of deeper understanding and reasoning. Our project page is at //github.com/OpenCausaLab/MORE.
As Artificial Intelligence (AI) becomes integral to software development, understanding the social and cooperative dynamics that affect AI-driven organizational change is important. Yet, despite AI's rapid progress and influence, the human and cooperative facets of these shifts in software organizations remain relatively less explored. This study uses Behavioral Software Engineering (BSE) as a lens to examine these often-overlooked dimensions of AI transformation. Through a qualitative approach involving ten semi-structured interviews across four organizations that are undergoing AI transformations, we performed a thematic analysis that revealed numerous sub-themes linked to twelve BSE concepts across individual, group, and organizational levels. Since the organizations are at an early stage of transformation we found more emphasis on the individual level. Our findings further reveal six key challenges tied to these BSE aspects that the organizations face during their AI transformation. Aligned with change management literature, we emphasize that effective communication, proactive leadership, and resistance management are essential for successful AI integration. However, we also identify ethical considerations as critical in the AI context-an area largely overlooked in previous research. Furthermore, a narrative analysis illustrates how different roles within an organization experience the AI transition in unique ways. These insights underscore that AI transformation extends beyond technical solutions; it requires a thoughtful approach that balances technological and human factors.
Tactile perception is essential for human interaction with the environment and is becoming increasingly crucial in robotics. Tactile sensors like the BioTac mimic human fingertips and provide detailed interaction data. Despite its utility in applications like slip detection and object identification, this sensor is now deprecated, making many existing valuable datasets obsolete. However, recreating similar datasets with newer sensor technologies is both tedious and time-consuming. Therefore, it is crucial to adapt these existing datasets for use with new setups and modalities. In response, we introduce ACROSS, a novel framework for translating data between tactile sensors by exploiting sensor deformation information. We demonstrate the approach by translating BioTac signals into the DIGIT sensor. Our framework consists of first converting the input signals into 3D deformation meshes. We then transition from the 3D deformation mesh of one sensor to the mesh of another, and finally convert the generated 3D deformation mesh into the corresponding output space. We demonstrate our approach to the most challenging problem of going from a low-dimensional tactile representation to a high-dimensional one. In particular, we transfer the tactile signals of a BioTac sensor to DIGIT tactile images. Our approach enables the continued use of valuable datasets and the exchange of data between groups with different setups.
Multimodal Large Language Models (MLLMs) have gained significant attention due to their impressive capabilities in multimodal understanding. However, existing methods rely heavily on extensive modal-specific pretraining and joint-modal tuning, leading to significant computational burdens when expanding to new modalities. In this paper, we propose PathWeave, a flexible and scalable framework with modal-Path sWitching and ExpAnsion abilities that enables MLLMs to continually EVolve on modalities for $\mathbb{X}$-modal reasoning. We leverage the concept of Continual Learning and develop an incremental training strategy atop pre-trained MLLMs, enabling their expansion to new modalities using uni-modal data, without executing joint-modal pretraining. In detail, a novel Adapter-in-Adapter (AnA) framework is introduced, in which uni-modal and cross-modal adapters are seamlessly integrated to facilitate efficient modality alignment and collaboration. Additionally, an MoE-based gating module is applied between two types of adapters to further enhance the multimodal interaction. To investigate the proposed method, we establish a challenging benchmark called Continual Learning of Modality (MCL), which consists of high-quality QA data from five distinct modalities: image, video, audio, depth and point cloud. Extensive experiments demonstrate the effectiveness of the proposed AnA framework on learning plasticity and memory stability during continual learning. Furthermore, PathWeave performs comparably to state-of-the-art MLLMs while concurrently reducing parameter training burdens by 98.73%. Our code locates at //github.com/JiazuoYu/PathWeave
This scoping review examines the current body of knowledge at the intersection of Generative Artificial Intelligence (GenAI) and Self-Directed Learning (SDL). By synthesising the findings from 18 studies published from 2020 to 2024 and following the PRISMA-SCR guidelines for scoping reviews, we developed four key themes. This includes GenAI as a Potential Enhancement for SDL, The Educator as a GenAI Guide, Personalisation of Learning, and Approaching with Caution. Our findings suggest that GenAI tools, including ChatGPT and other Large Language Models (LLMs) show promise in potentially supporting SDL through on-demand, personalised assistance. At the same time, the literature emphasises that educators are as important and central to the learning process as ever before, although their role may continue to shift as technologies develop. Our review reveals that there are still significant gaps in understanding the long-term impacts of GenAI on SDL outcomes, and there is a further need for longitudinal empirical studies that explore not only text-based chatbots but also emerging multimodal applications.
Reasoning is a fundamental aspect of human intelligence that plays a crucial role in activities such as problem solving, decision making, and critical thinking. In recent years, large language models (LLMs) have made significant progress in natural language processing, and there is observation that these models may exhibit reasoning abilities when they are sufficiently large. However, it is not yet clear to what extent LLMs are capable of reasoning. This paper provides a comprehensive overview of the current state of knowledge on reasoning in LLMs, including techniques for improving and eliciting reasoning in these models, methods and benchmarks for evaluating reasoning abilities, findings and implications of previous research in this field, and suggestions on future directions. Our aim is to provide a detailed and up-to-date review of this topic and stimulate meaningful discussion and future work.
In pace with developments in the research field of artificial intelligence, knowledge graphs (KGs) have attracted a surge of interest from both academia and industry. As a representation of semantic relations between entities, KGs have proven to be particularly relevant for natural language processing (NLP), experiencing a rapid spread and wide adoption within recent years. Given the increasing amount of research work in this area, several KG-related approaches have been surveyed in the NLP research community. However, a comprehensive study that categorizes established topics and reviews the maturity of individual research streams remains absent to this day. Contributing to closing this gap, we systematically analyzed 507 papers from the literature on KGs in NLP. Our survey encompasses a multifaceted review of tasks, research types, and contributions. As a result, we present a structured overview of the research landscape, provide a taxonomy of tasks, summarize our findings, and highlight directions for future work.
Besides entity-centric knowledge, usually organized as Knowledge Graph (KG), events are also an essential kind of knowledge in the world, which trigger the spring up of event-centric knowledge representation form like Event KG (EKG). It plays an increasingly important role in many machine learning and artificial intelligence applications, such as intelligent search, question-answering, recommendation, and text generation. This paper provides a comprehensive survey of EKG from history, ontology, instance, and application views. Specifically, to characterize EKG thoroughly, we focus on its history, definitions, schema induction, acquisition, related representative graphs/systems, and applications. The development processes and trends are studied therein. We further summarize perspective directions to facilitate future research on EKG.
UMAP (Uniform Manifold Approximation and Projection) is a novel manifold learning technique for dimension reduction. UMAP is constructed from a theoretical framework based in Riemannian geometry and algebraic topology. The result is a practical scalable algorithm that applies to real world data. The UMAP algorithm is competitive with t-SNE for visualization quality, and arguably preserves more of the global structure with superior run time performance. Furthermore, UMAP has no computational restrictions on embedding dimension, making it viable as a general purpose dimension reduction technique for machine learning.
ASR (automatic speech recognition) systems like Siri, Alexa, Google Voice or Cortana has become quite popular recently. One of the key techniques enabling the practical use of such systems in people's daily life is deep learning. Though deep learning in computer vision is known to be vulnerable to adversarial perturbations, little is known whether such perturbations are still valid on the practical speech recognition. In this paper, we not only demonstrate such attacks can happen in reality, but also show that the attacks can be systematically conducted. To minimize users' attention, we choose to embed the voice commands into a song, called CommandSong. In this way, the song carrying the command can spread through radio, TV or even any media player installed in the portable devices like smartphones, potentially impacting millions of users in long distance. In particular, we overcome two major challenges: minimizing the revision of a song in the process of embedding commands, and letting the CommandSong spread through the air without losing the voice "command". Our evaluation demonstrates that we can craft random songs to "carry" any commands and the modify is extremely difficult to be noticed. Specially, the physical attack that we play the CommandSongs over the air and record them can success with 94 percentage.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191