In recent years, semi-supervised learning (SSL) has gained significant attention due to its ability to leverage both labeled and unlabeled data to improve model performance, especially when labeled data is scarce. However, most current SSL methods rely on heuristics or predefined rules for generating pseudo-labels and leveraging unlabeled data. They are limited to exploiting loss functions and regularization methods within the standard norm. In this paper, we propose a novel Reinforcement Learning (RL) Guided SSL method, RLGSSL, that formulates SSL as a one-armed bandit problem and deploys an innovative RL loss based on weighted reward to adaptively guide the learning process of the prediction model. RLGSSL incorporates a carefully designed reward function that balances the use of labeled and unlabeled data to enhance generalization performance. A semi-supervised teacher-student framework is further deployed to increase the learning stability. We demonstrate the effectiveness of RLGSSL through extensive experiments on several benchmark datasets and show that our approach achieves consistent superior performance compared to state-of-the-art SSL methods.

In large language model training, input documents are typically concatenated together and then split into sequences of equal length to avoid padding tokens. Despite its efficiency, the concatenation approach compromises data integrity -- it inevitably breaks many documents into incomplete pieces, leading to excessive truncations that hinder the model from learning to compose logically coherent and factually consistent content that is grounded on the complete context. To address the issue, we propose Best-fit Packing, a scalable and efficient method that packs documents into training sequences through length-aware combinatorial optimization. Our method completely eliminates unnecessary truncations while retaining the same training efficiency as concatenation. Empirical results from both text and code pre-training show that our method achieves superior performance (e.g., relatively +4.7% on reading comprehension; +16.8% in context following; and +9.2% on program synthesis), and reduces closed-domain hallucination effectively by up to 58.3%.

Varied approaches for aligning language models have been proposed, including supervised fine-tuning, RLHF, and direct optimization methods such as DPO. Although DPO has rapidly gained popularity due to its straightforward training process and competitive results, there is an open question of whether there remain practical advantages of using a discriminator, like a reward model, to evaluate responses. We propose D2PO, discriminator-guided DPO, an approach for the online setting where preferences are being collected throughout learning. As we collect gold preferences, we use these not only to train our policy, but to train a discriminative response evaluation model to silver-label even more synthetic data for policy training. We explore this approach across a set of diverse tasks, including a realistic chat setting, we find that our approach leads to higher-quality outputs compared to DPO with the same data budget, and greater efficiency in terms of preference data requirements. Furthermore, we show conditions under which silver labeling is most helpful: it is most effective when training the policy with DPO, outperforming traditional PPO, and benefits from maintaining a separate discriminator from the policy model.

Large 2D vision-language models (2D-LLMs) have gained significant attention by bridging Large Language Models (LLMs) with images using a simple projector. Inspired by their success, large 3D point cloud-language models (3D-LLMs) also integrate point clouds into LLMs. However, directly aligning point clouds with LLM requires expensive training costs, typically in hundreds of GPU-hours on A100, which hinders the development of 3D-LLMs. In this paper, we introduce MiniGPT-3D, an efficient and powerful 3D-LLM that achieves multiple SOTA results while training for only 27 hours on one RTX 3090. Specifically, we propose to align 3D point clouds with LLMs using 2D priors from 2D-LLMs, which can leverage the similarity between 2D and 3D visual information. We introduce a novel four-stage training strategy for modality alignment in a cascaded way, and a mixture of query experts module to adaptively aggregate features with high efficiency. Moreover, we utilize parameter-efficient fine-tuning methods LoRA and Norm fine-tuning, resulting in only 47.8M learnable parameters, which is up to 260x fewer than existing methods. Extensive experiments show that MiniGPT-3D achieves SOTA on 3D object classification and captioning tasks, with significantly cheaper training costs. Notably, MiniGPT-3D gains an 8.12 increase on GPT-4 evaluation score for the challenging object captioning task compared to ShapeLLM-13B, while the latter costs 160 total GPU-hours on 8 A800. We are the first to explore the efficient 3D-LLM, offering new insights to the community. Code and weights are available at https://github.com/TangYuan96/MiniGPT-3D.

The rapid emergence of generative Language Models (LMs) has led to growing concern about the impacts that their unexamined adoption may have on the social well-being of diverse user groups. Meanwhile, LMs are increasingly being adopted in K-20 schools and one-on-one student settings with minimal investigation of potential harms associated with their deployment. Motivated in part by real-world/everyday use cases (e.g., an AI writing assistant) this paper explores the potential psychosocial harms of stories generated by five leading LMs in response to open-ended prompting. We extend findings of stereotyping harms analyzing a total of 150K 100-word stories related to student classroom interactions. Examining patterns in LM-generated character demographics and representational harms (i.e., erasure, subordination, and stereotyping) we highlight particularly egregious vignettes, illustrating the ways LM-generated outputs may influence the experiences of users with marginalized and minoritized identities, and emphasizing the need for a critical understanding of the psychosocial impacts of generative AI tools when deployed and utilized in diverse social contexts.

Attribution methods compute importance scores for input features to explain the output predictions of deep models. However, accurate assessment of attribution methods is challenged by the lack of benchmark fidelity for attributing model predictions. Moreover, other confounding factors in attribution estimation, including the setup choices of post-processing techniques and explained model predictions, further compromise the reliability of the evaluation. In this work, we first identify a set of fidelity criteria that reliable benchmarks for attribution methods are expected to fulfill, thereby facilitating a systematic assessment of attribution benchmarks. Next, we introduce a Backdoor-based eXplainable AI benchmark (BackX) that adheres to the desired fidelity criteria. We theoretically establish the superiority of our approach over the existing benchmarks for well-founded attribution evaluation. With extensive analysis, we also identify a setup for a consistent and fair benchmarking of attribution methods across different underlying methodologies. This setup is ultimately employed for a comprehensive comparison of existing methods using our BackX benchmark. Finally, our analysis also provides guidance for defending against backdoor attacks with the help of attribution methods.

Unmanned Aerial Vehicles (UAVs) have emerged as a transformative technology across diverse sectors, offering adaptable solutions to complex challenges in both military and civilian domains. Their expanding capabilities present a platform for further advancement by integrating cutting-edge computational tools like Artificial Intelligence (AI) and Machine Learning (ML) algorithms. These advancements have significantly impacted various facets of human life, fostering an era of unparalleled efficiency and convenience. Large Language Models (LLMs), a key component of AI, exhibit remarkable learning and adaptation capabilities within deployed environments, demonstrating an evolving form of intelligence with the potential to approach human-level proficiency. This work explores the significant potential of integrating UAVs and LLMs to propel the development of autonomous systems. We comprehensively review LLM architectures, evaluating their suitability for UAV integration. Additionally, we summarize the state-of-the-art LLM-based UAV architectures and identify novel opportunities for LLM embedding within UAV frameworks. Notably, we focus on leveraging LLMs to refine data analysis and decision-making processes, specifically for enhanced spectral sensing and sharing in UAV applications. Furthermore, we investigate how LLM integration expands the scope of existing UAV applications, enabling autonomous data processing, improved decision-making, and faster response times in emergency scenarios like disaster response and network restoration. Finally, we highlight crucial areas for future research that are critical for facilitating the effective integration of LLMs and UAVs.

In the pursuit of refining precise perception models for fully autonomous driving, continual online model training becomes essential. Federated Learning (FL) within vehicular networks offers an efficient mechanism for model training while preserving raw sensory data integrity. Yet, FL struggles with non-identically distributed data (e.g., quantity skew), leading to suboptimal convergence rates during model training. In previous work, we introduced FedLA, an innovative Label-Aware aggregation method addressing data heterogeneity in FL for generic scenarios. In this paper, we introduce FedProx+LA, a novel FL method building upon the state-of-the-art FedProx and FedLA to tackle data heterogeneity, which is specifically tailored for vehicular networks. We evaluate the efficacy of FedProx+LA in continuous online object detection model training. Through a comparative analysis against conventional and state-of-the-art methods, our findings reveal the superior convergence rate of FedProx+LA. Notably, if the label distribution is very heterogeneous, our FedProx+LA approach shows substantial improvements in detection performance compared to baseline methods, also outperforming our previous FedLA approach. Moreover, both FedLA and FedProx+LA increase convergence speed by 30% compared to baseline methods.

OpenAI's ChatGPT is facing serious competition, as the company's rival Anthropic brings its Claude chatbot to iPhones. Anthropic, led by a group of former OpenAI staff who quit over differences with chief executive Sam Altman, have a product that already beats ChatGPT on some measures of intelligence, and now wants to win over everyday users. "In today's world, smartphones are at the centre of how people interact with technology. To make Claude a true AI assistant, it's crucial that we meet users where they are – and in many cases, that's on their mobile devices," said Scott White at Anthropic. The third version of the Claude chatbot is offered direct to users on its website in three flavours: a speedy and simple model called "haiku", a slower and more powerful model called "sonnet", and, for paying customers only, the full "opus" system.

Every few years, an emerging technology shows up at the doorstep of schools and universities promising to transform education. Technologies and apps that include or are powered by generative artificial intelligence, also known as GenAI. These technologies are sold on the potential they hold for education. For example, Khan Academy's founder opened his 2023 Ted Talk by arguing that "we're at the cusp of using AI for probably the biggest positive transformation that education has ever seen." As optimistic as these visions of the future may be, the realities of educational technology over the past few decades have not lived up to their promises.