Large Language Models (LLMs) have demonstrated significant potential in causal discovery tasks by utilizing their vast expert knowledge from extensive text corpora. However, the multi-agent capabilities of LLMs in causal discovery remain underexplored. This paper introduces a general framework to investigate this potential. The first is the Meta Agents Model, which relies exclusively on reasoning and discussions among LLM agents to conduct causal discovery. The second is the Coding Agents Model, which leverages the agents' ability to plan, write, and execute code, utilizing advanced statistical libraries for causal discovery. The third is the Hybrid Model, which integrates both the Meta Agents Model and Coding Agents Model approaches, combining the statistical analysis and reasoning skills of multiple agents. Our proposed framework shows promising results by effectively utilizing LLMs' expert knowledge, reasoning capabilities, multi-agent cooperation, and statistical causal methods. By exploring the multi-agent potential of LLMs, we aim to establish a foundation for further research in utilizing LLMs multi-agent for solving causal-related problems.

Scarlett Johansson is speaking out about the reasons she turned down the job of voicing OpenAI's chatbot. Last year, OpenAI CEO Sam Altman reached out to the 39-year-old actress about potentially hiring her to voice the ChatGPT 4.0 system. In an interview with The New York Times, Johansson, who voiced the character of Samantha, an artificial intelligence virtual assistant in the 2013 film "Her," recalled that she said, "No, thank you. Not for me," when Altman approached her about the gig. "I felt I did not want to be at the forefront of that," Johansson told the Times.

Recent advances in multimodal LLMs, have led to several video-text models being proposed for critical video-related tasks. However, most of the previous works support visual input only, essentially muting the audio signal in the video. Few models that support both audio and visual input, are not explicitly trained on audio data. Hence, the effect of audio towards video understanding is largely unexplored. To this end, we propose a model architecture that handles audio-visual inputs explicitly. We train our model with both audio and visual data from a video instruction-tuning dataset. Comparison with vision-only baselines, and other audiovisual models showcase that training on audio data indeed leads to improved grounding of responses. Figure 1: An example of improved grounding in the For better evaluation of audio-visual video-text LLM outputs, due to the additional audio models, we also release a human-annotated signal as input.

We initiate a formal investigation into the design and analysis of LLM-based algorithms, i.e. algorithms that contain one or multiple calls of large language models (LLMs) as sub-routines and critically rely on the capabilities of LLMs. While LLM-based algorithms, ranging from basic LLM calls with prompt engineering to complicated LLM-powered agent systems and compound AI systems, have achieved remarkable empirical success, the design and optimization of them have mostly relied on heuristics and trial-and-errors, which is largely due to a lack of formal and analytical study for these algorithms. To fill this gap, we start by identifying the computational-graph representation of LLM-based algorithms, the design principle of task decomposition, and some key abstractions, which then facilitate our formal analysis for the accuracy and efficiency of LLM-based algorithms, despite the black-box nature of LLMs. We further consider parallel decomposition for a case study, providing extensive analytical and empirical study for four concrete examples of this pattern. Our proposed framework holds promise for advancing LLM-based algorithms, by revealing the reasons behind curious empirical phenomena, guiding the choices of hyperparameters, predicting the empirical performance of algorithms, and inspiring new algorithm design. To promote further study of LLM-based algorithms, we release our source code at https://github.com/modelscope/agentscope/tree/main/examples/paper_llm_based_algorithm.

This paper introduces Golden-Retriever, designed to efficiently navigate vast industrial knowledge bases, overcoming challenges in traditional LLM fine-tuning and RAG frameworks with domain-specific jargon and context interpretation. Golden-Retriever incorporates a reflection-based question augmentation step before document retrieval, which involves identifying jargon, clarifying its meaning based on context, and augmenting the question accordingly. Specifically, our method extracts and lists all jargon and abbreviations in the input question, determines the context against a pre-defined list, and queries a jargon dictionary for extended definitions and descriptions. This comprehensive augmentation ensures the RAG framework retrieves the most relevant documents by providing clear context and resolving ambiguities, significantly improving retrieval accuracy. Evaluations using three open-source LLMs on a domain-specific question-answer dataset demonstrate Golden-Retriever's superior performance, providing a robust solution for efficiently integrating and querying industrial knowledge bases.

The application of large-scale models in medical image segmentation demands substantial quantities of meticulously annotated data curated by experts along with high computational resources, both of which are challenges in resource-poor settings. In this study, we present the Medical Segment Anything Model with Galore MedSAGa where we adopt the Segment Anything Model (SAM) to achieve memory-efficient, few-shot medical image segmentation by applying Gradient Low-Rank Projection GaLore to the parameters of the image encoder of SAM. Meanwhile, the weights of the prompt encoder and mask decoder undergo full parameter fine-tuning using standard optimizers. We further assess MedSAGa's few-shot learning capabilities, reporting on its memory efficiency and segmentation performance across multiple standard medical image segmentation datasets. We compare it with several baseline models, including LoRA fine-tuned SAM (SAMed) and DAE-Former. Experiments across multiple datasets and these baseline models with different number of images for fine tuning demonstrated that the GPU memory consumption of MedSAGa is significantly less than that of the baseline models, achieving an average memory efficiency of 66% more than current state-of-the-art (SOTA) models for medical image segmentation. The combination of substantially lower memory requirements and comparable to SOTA results in few-shot learning for medical image segmentation positions MedSAGa as an optimal solution for deployment in resource-constrained settings.

Despite the proven utility of large language models (LLMs) in real-world applications, there remains a lack of understanding regarding how they leverage their large-scale pretraining text corpora to achieve such capabilities. In this work, we investigate the interplay between generalization and memorization in pretrained LLMs at scale, through a comprehensive $n$-gram analysis of their training data. Our experiments focus on three general task types: translation, question-answering, and multiple-choice reasoning. With various sizes of open-source LLMs and their pretraining corpora, we observe that as the model size increases, the task-relevant $n$-gram pair data becomes increasingly important, leading to improved task performance, decreased memorization, stronger generalization, and emergent abilities. Our results support the hypothesis that LLMs' capabilities emerge from a delicate balance of memorization and generalization with sufficient task-related pretraining data, and point the way to larger-scale analyses that could further improve our understanding of these models.

While finetuning language models from pairwise preferences has proven remarkably effective, the underspecified nature of natural language presents critical challenges. Direct preference feedback is uninterpretable, difficult to provide where multidimensional criteria may apply, and often inconsistent, either because it is based on incomplete instructions or provided by diverse principals. To address these challenges, we consider the two-step preference modeling procedure that first resolves the under-specification by selecting a context, and then evaluates preference with respect to the chosen context. We decompose reward modeling error according to these two steps, which suggests that supervising context in addition to context-specific preference may be a viable approach to aligning models with diverse human preferences. For this to work, the ability of models to evaluate context-specific preference is critical. To this end, we contribute context-conditioned preference datasets and accompanying experiments that investigate the ability of language models to evaluate context-specific preference. We use our datasets to (1) show that existing preference models benefit from, but fail to fully consider, added context, (2) finetune a context-aware reward model with context-specific performance exceeding that of GPT-4 and Llama 3 70B on tested datasets, and (3) investigate the value of context-aware preference modeling.

Imagine there is a disruption in train 1 near Times Square metro station. You try to find an alternative subway route to the JFK airport on Google Maps, but the app fails to provide a suitable recommendation that takes into account the disruption and your preferences to avoid crowded stations. We find that in many such situations, current navigation apps may fall short and fail to give a reasonable recommendation. To fill this gap, in this paper, we develop a prototype, TraveLLM, to plan routing of public transit in face of disruption that relies on Large Language Models (LLMs). LLMs have shown remarkable capabilities in reasoning and planning across various domains. Here we hope to investigate the potential of LLMs that lies in incorporating multi-modal user-specific queries and constraints into public transit route recommendations. Various test cases are designed under different scenarios, including varying weather conditions, emergency events, and the introduction of new transportation services. We then compare the performance of state-of-the-art LLMs, including GPT-4, Claude 3 and Gemini, in generating accurate routes. Our comparative analysis demonstrates the effectiveness of LLMs, particularly GPT-4 in providing navigation plans. Our findings hold the potential for LLMs to enhance existing navigation systems and provide a more flexible and intelligent method for addressing diverse user needs in face of disruptions.

In this study, we explore the proactive ability of LLMs to seek user support, using text-to-SQL generation as a case study. We propose metrics to evaluate the trade-off between performance improvements and user burden, and investigate whether LLMs can determine when to request help and examine their performance with varying levels of information availability. Our experiments reveal that without external feedback, many LLMs struggle to recognize their need for additional support. Our findings highlight the importance of external signals and provide insights for future research on improving support-seeking strategies.