We introduce VideoPrism, a general-purpose video encoder that tackles diverse video understanding tasks with a single frozen model. We pretrain VideoPrism on a heterogeneous corpus containing 36M high-quality video-caption pairs and 582M video clips with noisy parallel text (e.g., ASR transcripts). The pretraining approach improves upon masked autoencoding by global-local distillation of semantic video embeddings and a token shuffling scheme, enabling VideoPrism to focus primarily on the video modality while leveraging the invaluable text associated with videos. We extensively test VideoPrism on four broad groups of video understanding tasks, from web video question answering to CV for science, achieving state-of-the-art performance on 31 out of 33 video understanding benchmarks.

As the use of Large Language Models (LLMs) becomes more widespread, understanding their self-evaluation of confidence in generated responses becomes increasingly important as it is integral to the reliability of the output of these models. We introduce the concept of Confidence-Probability Alignment, that connects an LLM's internal confidence, quantified by token probabilities, to the confidence conveyed in the model's response when explicitly asked about its certainty. Using various datasets and prompting techniques that encourage model introspection, we probe the alignment between models' internal and expressed confidence. These techniques encompass using structured evaluation scales to rate confidence, including answer options when prompting, and eliciting the model's confidence level for outputs it does not recognize as its own. Notably, among the models analyzed, OpenAI's GPT-4 showed the strongest confidence-probability alignment, with an average Spearman's $\hat{\rho}$ of 0.42, across a wide range of tasks. Our work contributes to the ongoing efforts to facilitate risk assessment in the application of LLMs and to further our understanding of model trustworthiness.

This paper critically discusses how generative artificial intelligence (GenAI) might impose Western ideologies on non-Western societies, perpetuating digital neocolonialism in education through its inherent biases. It further suggests strategies for local and global stakeholders to mitigate these effects. Our discussions demonstrated that GenAI can foster cultural imperialism by generating content that primarily incorporates cultural references and examples relevant to Western students, thereby alienating students from non-Western backgrounds. Also, the predominant use of Western languages by GenAI can marginalize non-dominant languages, making educational content less accessible to speakers of indigenous languages and potentially impacting their ability to learn in their first language. Additionally, GenAI often generates content and curricula that reflect the perspectives of technologically dominant countries, overshadowing marginalized indigenous knowledge and practices. Moreover, the cost of access to GenAI intensifies educational inequality and the control of GenAI data could lead to commercial exploitation without benefiting local students and their communities. We propose human-centric reforms to prioritize cultural diversity and equity in GenAI development; a liberatory design to empower educators and students to identify and dismantle the oppressive structures within GenAI applications; foresight by design to create an adjustable GenAI system to meet future educational needs; and finally, effective prompting skills to reduce the retrieval of neocolonial outputs.

Large language models (LLMs) have significantly transformed the educational landscape. As current plagiarism detection tools struggle to keep pace with LLMs' rapid advancements, the educational community faces the challenge of assessing students' true problem-solving abilities in the presence of LLMs. In this work, we explore a new paradigm for ensuring fair evaluation -- generating adversarial examples which preserve the structure and difficulty of the original questions aimed for assessment, but are unsolvable by LLMs. Focusing on the domain of math word problems, we leverage abstract syntax trees to structurally generate adversarial examples that cause LLMs to produce incorrect answers by simply editing the numeric values in the problems. We conduct experiments on various open- and closed-source LLMs, quantitatively and qualitatively demonstrating that our method significantly degrades their math problem-solving ability. We identify shared vulnerabilities among LLMs and propose a cost-effective approach to attack high-cost models. Additionally, we conduct automatic analysis to investigate the cause of failure, providing further insights into the limitations of LLMs.

Large language models (LLMs) often struggle to provide up-to-date information due to their one-time training and the constantly evolving nature of the world. To keep LLMs current, existing approaches typically involve continued pre-training on new documents. However, they frequently face difficulties in extracting stored knowledge. Motivated by the remarkable success of the Feynman Technique in efficient human learning, we introduce Self-Tuning, a learning framework aimed at improving an LLM's ability to effectively acquire new knowledge from raw documents through self-teaching. Specifically, we develop a Self-Teaching strategy that augments the documents with a set of knowledge-intensive tasks created in a self-supervised manner, focusing on three crucial aspects: memorization, comprehension, and self-reflection. In addition, we introduce three Wiki-Newpages-2023-QA datasets to facilitate an in-depth analysis of an LLM's knowledge acquisition ability concerning memorization, extraction, and reasoning. Extensive experimental results on Llama2 family models reveal that Self-Tuning consistently exhibits superior performance across all knowledge acquisition tasks and excels in preserving previous knowledge.

Previous studies on framing have relied on manual analysis or fine-tuning models with limited annotated datasets. However, pre-trained models, with their diverse training backgrounds, offer a promising alternative. This paper presents a comprehensive analysis of GPT-4, GPT-3.5 Turbo, and FLAN-T5 models in detecting framing in news headlines. We evaluated these models in various scenarios: zero-shot, few-shot with in-domain examples, cross-domain examples, and settings where models explain their predictions. Our results show that explainable predictions lead to more reliable outcomes. GPT-4 performed exceptionally well in few-shot settings but often misinterpreted emotional language as framing, highlighting a significant challenge. Additionally, the results suggest that consistent predictions across multiple models could help identify potential annotation inaccuracies in datasets. Finally, we propose a new small dataset for real-world evaluation on headlines from a diverse set of topics.

This spring, the Los Angeles Unified School District--the second-largest public school district in the United States--introduced students and parents to a new "educational friend" named Ed. A learning platform that includes a chatbot represented by a small illustration of a smiling sun, Ed is being tested in 100 schools within the district and is accessible at all hours through a website. It can answer questions about a child's courses, grades, and attendance, and point users to optional activities. As Superintendent Alberto M. Carvalho put it to me, "AI is here to stay. If you don't master it, it will master you." Carvalho says he wants to empower teachers and students to learn to use AI safely.

Training corpuses for vision language models (VLMs) typically lack sufficient amounts of decision-centric data. This renders off-the-shelf VLMs sub-optimal for decision-making tasks such as in-the-wild device control through graphical user interfaces (GUIs). While training with static demonstrations has shown some promise, we show that such methods fall short for controlling real GUIs due to their failure to deal with real-world stochasticity and non-stationarity not captured in static observational data. This paper introduces a novel autonomous RL approach, called DigiRL, for training in-the-wild device control agents through fine-tuning a pre-trained VLM in two stages: offline RL to initialize the model, followed by offline-to-online RL. To do this, we build a scalable and parallelizable Android learning environment equipped with a VLM-based evaluator and develop a simple yet effective RL approach for learning in this domain. Our approach runs advantage-weighted RL with advantage estimators enhanced to account for stochasticity along with an automatic curriculum for deriving maximal learning signal. We demonstrate the effectiveness of DigiRL using the Android-in-the-Wild (AitW) dataset, where our 1.3B VLM trained with RL achieves a 49.5% absolute improvement -- from 17.7 to 67.2% success rate -- over supervised fine-tuning with static human demonstration data. These results significantly surpass not only the prior best agents, including AppAgent with GPT-4V (8.3% success rate) and the 17B CogAgent trained with AitW data (38.5%), but also the prior best autonomous RL approach based on filtered behavior cloning (57.8%), thereby establishing a new state-of-the-art for digital agents for in-the-wild device control.

Artificial agents, particularly humanoid robots, interact with their environment, objects, and people using cameras, actuators, and physical presence. Their communication methods are often pre-programmed, limiting their actions and interactions. Our research explores acquiring non-verbal communication skills through learning from demonstrations, with potential applications in sign language comprehension and expression. In particular, we focus on imitation learning for artificial agents, exemplified by teaching a simulated humanoid American Sign Language. We use computer vision and deep learning to extract information from videos, and reinforcement learning to enable the agent to replicate observed actions. Compared to other methods, our approach eliminates the need for additional hardware to acquire information. We demonstrate how the combination of these different techniques offers a viable way to learn sign language. Our methodology successfully teaches 5 different signs involving the upper body (i.e., arms and hands). This research paves the way for advanced communication skills in artificial agents.

Knowledge distillation (KD) is a technique used to transfer knowledge from an overparameterized teacher network to a less-parameterized student network, thereby minimizing the incurred performance loss. KD methods can be categorized into offline and online approaches. Offline KD leverages a powerful pretrained teacher network, while online KD allows the teacher network to be adjusted dynamically to enhance the learning effectiveness of the student network. Recently, it has been discovered that sharing the classifier of the teacher network can significantly boost the performance of the student network with only a minimal increase in the number of network parameters. Building on these insights, we propose adaptive teaching with a shared classifier (ATSC). In ATSC, the pretrained teacher network self-adjusts to better align with the learning needs of the student network based on its capabilities, and the student network benefits from the shared classifier, enhancing its performance. Additionally, we extend ATSC to environments with multiple teachers. We conduct extensive experiments, demonstrating the effectiveness of the proposed KD method. Our approach achieves state-of-the-art results on the CIFAR-100 and ImageNet datasets in both single-teacher and multiteacher scenarios, with only a modest increase in the number of required model parameters.