Open-source LLMs have shown great potential as fine-tuned chatbots, and demonstrate robust abilities in reasoning and surpass many existing benchmarks. Retrieval-Augmented Generation (RAG) is a technique for improving the performance of LLMs on tasks that the models weren't explicitly trained on, by leveraging external knowledge databases. Numerous studies have demonstrated the effectiveness of RAG to more successfully accomplish downstream tasks when using vector datasets that consist of relevant background information. It has been implicitly assumed by those in the field that if adversarial background information is utilized in this context, that the success of using a RAG-based approach would be nonexistent or even negatively impact the results. To address this assumption, we tested several open-source LLMs on the ability of RAG to improve their success in answering multiple-choice questions (MCQ) in the medical subspecialty field of Nephrology. Unlike previous studies, we examined the effect of RAG in utilizing both relevant and adversarial background databases. We set up several open-source LLMs, including Llama 3, Phi-3, Mixtral 8x7b, Zephyr$\beta$, and Gemma 7B Instruct, in a zero-shot RAG pipeline. As adversarial sources of information, text from the Bible and a Random Words generated database were used for comparison. Our data show that most of the open-source LLMs improve their multiple-choice test-taking success as expected when incorporating relevant information vector databases. Surprisingly however, adversarial Bible text significantly improved the success of many LLMs and even random word text improved test taking ability of some of the models. In summary, our results demonstrate for the first time the countertintuitive ability of adversarial information datasets to improve the RAG-based LLM success.

Large language models (LLMs) targeting different deployment scales and sizes are currently produced by training each variant from scratch; this is extremely compute-intensive. In this paper, we investigate if pruning an existing LLM and then re-training it with a fraction (<3%) of the original training data can be a suitable alternative to repeated, full retraining. To this end, we develop a set of practical and effective compression best practices for LLMs that combine depth, width, attention and MLP pruning with knowledge distillation-based retraining; we arrive at these best practices through a detailed empirical exploration of pruning strategies for each axis, methods to combine axes, distillation strategies, and search techniques for arriving at optimal compressed architectures. We use this guide to compress the Nemotron-4 family of LLMs by a factor of 2-4x, and compare their performance to similarly-sized models on a variety of language modeling tasks. Deriving 8B and 4B models from an already pretrained 15B model using our approach requires up to 40x fewer training tokens per model compared to training from scratch; this results in compute cost savings of 1.8x for training the full model family (15B, 8B, and 4B). Minitron models exhibit up to a 16% improvement in MMLU scores compared to training from scratch, perform comparably to other community models such as Mistral 7B, Gemma 7B and Llama-3 8B, and outperform state-of-the-art compression techniques from the literature. We have open-sourced Minitron model weights on Huggingface, with corresponding supplementary material including example code available on GitHub.

OpenAI on Thursday released a smaller and more affordable version of its flagship large language model that powers ChatGPT. The new model, called GPT-4o mini, will reportedly cost developers 60 percent less to build AI-powered apps and services with as compared to GPT-3.5 Turbo, Open's smallest model until today. But the big news here is for consumers. GPT-4o mini will replace GPT-3.5 Turbo for free users of ChatGPT starting today -- which means that your baseline ChatGPT experience will improve significantly. OpenAI claimed that GPT-4o mini achieved an 82 percent score on an industry benchmark called the MMLU, which stands for Measuring Massive Multitask Language Understanding, and includes 16,000 multiple-choice questions across 57 academic subjects.

I am by no means a skilled coder, but thanks to a free program called SWE-agent, I was just able to debug and fix a gnarly problem involving a misnamed file within different code repositories on the software-hosting site GitHub. I pointed SWE-agent at an issue on GitHub and watched as it went through the code and reasoned about what might be wrong. It correctly determined that the root cause of the bug was a line that pointed to the wrong location for a file, then navigated through the project, located the file, and amended the code so that everything ran properly. It's the kind of thing that an inexperienced developer (such as myself) might spend hours trying to debug. Many coders already use artificial intelligence to write software more quickly.

OpenAI today announced a cut-price "mini" model that it says will allow more companies and programs to tap into its artificial intelligence. The new model, called GPT-4o mini and available starting today, is 60 percent cheaper than OpenAI's most inexpensive existing model while offering higher performance, the company says. OpenAI characterizes the move as part of an effort to make AI "as broadly accessible as possible," but it also reflects growing competition among AI cloud providers as well as rising interest in small and free open source AI models. Meta is expected to debut the largest version of its very capable free offering, Llama 3, next week. "The whole point of OpenAI is to build and distribute AI safely and make it broadly accessible," Olivier Godement, a product manager at OpenAI responsible for the new model, tells WIRED.

Today, you can ask Microsoft Copilot to draw an AI-generated image of a strawberry pie for free, right from within Windows. But if you want to do the same thing in Microsoft Word, it will cost you at least 26 dollars per month. Does that make sense to you? AI art has existed within Microsoft's ecosystem since 2022, when Microsoft released its first beta of Microsoft Bing Image Creator. That later became Microsoft Designer, the design tool that Microsoft developed to compete against Canva. This week, Microsoft said that Designer's AI art capabilities would roll out in several of its apps and services, including Photos, Word, and PowerPoint. There's a catch, though: Those capabilities don't come for free.

Embodied Artificial Intelligence (Embodied AI) is crucial for achieving Artificial General Intelligence (AGI) and serves as a foundation for various applications that bridge cyberspace and the physical world. Recently, the emergence of Multi-modal Large Models (MLMs) and World Models (WMs) have attracted significant attention due to their remarkable perception, interaction, and reasoning capabilities, making them a promising architecture for the brain of embodied agents. However, there is no comprehensive survey for Embodied AI in the era of MLMs. In this survey, we give a comprehensive exploration of the latest advancements in Embodied AI. Our analysis firstly navigates through the forefront of representative works of embodied robots and simulators, to fully understand the research focuses and their limitations. Then, we analyze four main research targets: 1) embodied perception, 2) embodied interaction, 3) embodied agent, and 4) sim-to-real adaptation, covering the state-of-the-art methods, essential paradigms, and comprehensive datasets. Additionally, we explore the complexities of MLMs in virtual and real embodied agents, highlighting their significance in facilitating interactions in dynamic digital and physical environments. Finally, we summarize the challenges and limitations of embodied AI and discuss their potential future directions. We hope this survey will serve as a foundational reference for the research community and inspire continued innovation. The associated project can be found at https://github.com/HCPLab-SYSU/Embodied_AI_Paper_List.

Techniques that learn improved representations via offline data or self-supervised objectives have shown impressive results in traditional reinforcement learning (RL). Nevertheless, it is unclear how improved representation learning can benefit reinforcement learning from human feedback (RLHF) on language models (LMs). In this work, we propose training reward models (RMs) in a contrastive, $\textit{goal-conditioned}$ fashion by increasing the representation similarity of future states along sampled preferred trajectories and decreasing the similarity along randomly sampled dispreferred trajectories. This objective significantly improves RM performance by up to 0.09 AUROC across challenging benchmarks, such as MATH and GSM8k. These findings extend to general alignment as well -- on the Helpful-Harmless dataset, we observe $2.3\%$ increase in accuracy. Beyond improving reward model performance, we show this way of training RM representations enables improved $\textit{steerability}$ because it allows us to evaluate the likelihood of an action achieving a particular goal-state (e.g., whether a solution is correct or helpful). Leveraging this insight, we find that we can filter up to $55\%$ of generated tokens during majority voting by discarding trajectories likely to end up in an "incorrect" state, which leads to significant cost savings. We additionally find that these representations can perform fine-grained control by conditioning on desired future goal-states. For example, we show that steering a Llama 3 model towards helpful generations with our approach improves helpfulness by $9.6\%$ over a supervised-fine-tuning trained baseline. Similarly, steering the model towards complex generations improves complexity by $21.6\%$ over the baseline. Overall, we find that training RMs in this contrastive, goal-conditioned fashion significantly improves performance and enables model steerability.

SpeechBrain is an open-source Conversational AI toolkit based on PyTorch, focused particularly on speech processing tasks such as speech recognition, speech enhancement, speaker recognition, text-to-speech, and much more. It promotes transparency and replicability by releasing both the pre-trained models and the complete "recipes" of code and algorithms required for training them. This paper presents SpeechBrain 1.0, a significant milestone in the evolution of the toolkit, which now has over 200 recipes for speech, audio, and language processing tasks, and more than 100 models available on Hugging Face. SpeechBrain 1.0 introduces new technologies to support diverse learning modalities, Large Language Model (LLM) integration, and advanced decoding strategies, along with novel models, tasks, and modalities. It also includes a new benchmark repository, offering researchers a unified platform for evaluating models across diverse tasks.

The field of medical diagnosis has undergone a significant transformation with the advent of large language models (LLMs), yet the challenges of interpretability within these models remain largely unaddressed. This study introduces Chain-of-Diagnosis (CoD) to enhance the interpretability of LLM-based medical diagnostics. CoD transforms the diagnostic process into a diagnostic chain that mirrors a physician's thought process, providing a transparent reasoning pathway. Additionally, CoD outputs the disease confidence distribution to ensure transparency in decisionmaking. This interpretability makes model diagnostics controllable and aids in identifying critical symptoms for inquiry through the entropy reduction of confidences. With CoD, we developed DiagnosisGPT, capable of diagnosing 9,604 diseases. Experimental results demonstrate that DiagnosisGPT outperforms other LLMs on diagnostic benchmarks. Moreover, DiagnosisGPT provides interpretability while ensuring controllability in diagnostic rigor.