Large Language Models (LLMs) have the unique capability to understand and generate human-like text from input queries. When fine-tuned, these models show enhanced performance on domain-specific queries. OpenAI highlights the process of fine-tuning, stating: "To fine-tune a model, you are required to provide at least 10 examples. We typically see clear improvements from fine-tuning on 50 to 100 training examples, but the right number varies greatly based on the exact use case." This study extends this concept to the integration of LLMs within Retrieval-Augmented Generation (RAG) pipelines, which aim to improve accuracy and relevance by leveraging external corpus data for information retrieval. However, RAG's promise of delivering optimal responses often falls short in complex query scenarios. This study aims to specifically examine the effects of fine-tuning LLMs on their ability to extract and integrate contextual data to enhance the performance of RAG systems across multiple domains. We evaluate the impact of fine-tuning on the LLMs' capacity for data extraction and contextual understanding by comparing the accuracy and completeness of fine-tuned models against baseline performances across datasets from multiple domains. Our findings indicate that fine-tuning resulted in a decline in performance compared to the baseline models, contrary to the improvements observed in standalone LLM applications as suggested by OpenAI. This study highlights the need for vigorous investigation and validation of fine-tuned models for domain-specific tasks.

In this paper, we will present SketchQL, a video database management system (VDBMS) for retrieving video moments with a sketch-based query interface. This novel interface allows users to specify object trajectory events with simple mouse drag-and-drop operations. Users can use trajectories of single objects as building blocks to compose complex events. Using a pre-trained model that encodes trajectory similarity, SketchQL achieves zero-shot video moments retrieval by performing similarity searches over the video to identify clips that are the most similar to the visual query. In this demonstration, we introduce the graphic user interface of SketchQL and detail its functionalities and interaction mechanisms. We also demonstrate the end-to-end usage of SketchQL from query composition to video moments retrieval using real-world scenarios.

Dual-arm robots offer enhanced versatility and efficiency over single-arm counterparts by enabling concurrent manipulation of multiple objects or cooperative execution of tasks using both arms. However, effectively coordinating the two arms for complex long-horizon tasks remains a significant challenge. Existing task planning methods predominantly focus on single-arm robots or rely on predefined bimanual operations, failing to fully leverage the capabilities of dual-arm systems. To address this limitation, we introduce DAG-Plan, a structured task planning framework tailored for dual-arm robots. DAG-Plan harnesses large language models (LLMs) to decompose intricate tasks into actionable sub-tasks represented as nodes within a directed acyclic graph (DAG). Critically, DAG-Plan dynamically assigns these sub-tasks to the appropriate arm based on real-time environmental observations, enabling parallel and adaptive execution. We evaluate DAG-Plan on the novel Dual-Arm Kitchen Benchmark, comprising 9 sequential tasks with 78 sub-tasks and 26 objects. Extensive experiments demonstrate the superiority of DAG-Plan over directly using LLM to generate plans, achieving nearly 50% higher efficiency compared to the single-arm task planning baseline and nearly double the success rate of the dual-arm task planning baseline.

Vision-and-language navigation (VLN) stands as a key research problem of Embodied AI, aiming at enabling agents to navigate in unseen environments following linguistic instructions. In this field, generalization is a long-standing challenge, either to out-of-distribution scenes or from Sim to Real. In this paper, we propose NaVid, a video-based large vision language model (VLM), to mitigate such a generalization gap. NaVid makes the first endeavor to showcase the capability of VLMs to achieve state-of-the-art level navigation performance without any maps, odometers, or depth inputs. Following human instruction, NaVid only requires an on-the-fly video stream from a monocular RGB camera equipped on the robot to output the next-step action. Our formulation mimics how humans navigate and naturally gets rid of the problems introduced by odometer noises, and the Sim2Real gaps from map or depth inputs. Moreover, our video-based approach can effectively encode the historical observations of robots as spatio-temporal contexts for decision making and instruction following. We train NaVid with 510k navigation samples collected from continuous environments, including action-planning and instruction-reasoning samples, along with 763k large-scale web data. Extensive experiments show that NaVid achieves state-of-the-art performance in simulation environments and the real world, demonstrating superior cross-dataset and Sim2Real transfer. We thus believe our proposed VLM approach plans the next step for not only the navigation agents but also this research field.

In-context learning (ICL) facilitates Large Language Models (LLMs) exhibiting emergent ability on downstream tasks without updating billions of parameters. However, in the area of multi-modal Large Language Models (MLLMs), two problems hinder the application of multi-modal ICL: (1) Most primary MLLMs are only trained on single-image datasets, making them unable to read multi-modal demonstrations. (2) With the demonstrations increasing, thousands of visual tokens highly challenge hardware and degrade ICL performance. During preliminary explorations, we discovered that the inner LLM tends to focus more on the linguistic modality within multi-modal demonstrations to generate responses. Therefore, we propose a general and light-weighted framework \textbf{AIM} to tackle the mentioned problems through \textbf{A}ggregating \textbf{I}mage information of \textbf{M}ultimodal demonstrations to the dense latent space of the corresponding linguistic part. Specifically, AIM first uses the frozen backbone MLLM to read each image-text demonstration and extracts the vector representations on top of the text. These vectors naturally fuse the information of the image-text pair, and AIM transforms them into fused virtual tokens acceptable for the inner LLM via a trainable projection layer. Ultimately, these fused tokens function as variants of multi-modal demonstrations, fed into the MLLM to direct its response to the current query as usual. Because these fused tokens stem from the textual component of the image-text pair, a multi-modal demonstration is nearly reduced to a pure textual demonstration, thus seamlessly applying to any MLLMs. With its de facto MLLM frozen, AIM is parameter-efficient and we train it on public multi-modal web corpora which have nothing to do with downstream test tasks.

If you want evidence of Microsoft's progress towards its environmental "moonshot" goal, then look closer to earth: at a building site on a west London industrial estate. The company's Park Royal datacentre is part of its commitment to drive the expansion of artificial intelligence (AI), but that ambition is jarring with its target of being carbon negative by 2030. Microsoft says the centre will be run fully on renewable energy. However, the construction of datacentres and the servers they are filled with means that the company's scope 3 emissions – such as CO2 related to the materials in its buildings and the electricity people consume when using products such as Xbox – are more than 30% above their 2020 level. As a result, the company is exceeding its overall emissions target by roughly the same rate.

The visual appeal of urban environments significantly impacts residents' satisfaction with their living spaces and their overall mood, which in turn, affects their health and well-being. Given the resource-intensive nature of gathering evaluations on urban visual appeal through surveys or inquiries from residents, there is a constant quest for automated solutions to streamline this process and support spatial planning. In this study, we applied an off-the-shelf AI model to automate the analysis of urban visual appeal, using over 1,800 Google Street View images of Helsinki, Finland. By incorporating the GPT-4 model with specified criteria, we assessed these images. Simultaneously, 24 participants were asked to rate the images. Our results demonstrated a strong alignment between GPT-4 and participant ratings, although geographic disparities were noted. Specifically, GPT-4 showed a preference for suburban areas with significant greenery, contrasting with participants who found these areas less appealing. Conversely, in the city centre and densely populated urban regions of Helsinki, GPT-4 assigned lower visual appeal scores than participant ratings. While there was general agreement between AI and human assessments across various locations, GPT-4 struggled to incorporate contextual nuances into its ratings, unlike participants, who considered both context and features of the urban environment. The study suggests that leveraging AI models like GPT-4 allows spatial planners to gather insights into the visual appeal of different areas efficiently, aiding decisions that enhance residents' and travellers' satisfaction and mental health. Although AI models provide valuable insights, human perspectives are essential for a comprehensive understanding of urban visual appeal. This will ensure that planning and design decisions promote healthy living environments effectively.

As the parameter size of large language models (LLMs) continues to expand, the need for a large memory footprint and high communication bandwidth have become significant bottlenecks for the training and inference of LLMs. To mitigate these bottlenecks, various tensor compression techniques have been proposed to reduce the data size, thereby alleviating memory requirements and communication pressure. Our research found that video codecs, despite being originally designed for compressing videos, show excellent efficiency when compressing various types of tensors. We demonstrate that video codecs can be versatile and general-purpose tensor codecs while achieving the state-of-the-art compression efficiency in various tasks. We further make use of the hardware video encoding and decoding module available on GPUs to create a framework capable of both inference and training with video codecs repurposed as tensor codecs. This greatly reduces the requirement for memory capacity and communication bandwidth, enabling training and inference of large models on consumer-grade GPUs.

Bridging the significant gap between large language model's English and non-English performance presents a great challenge. While some previous studies attempt to mitigate this gap with translated training data, the recently proposed question alignment approach leverages the model's English expertise to improve multilingual performance with minimum usage of expensive, error-prone translation. In this paper, we explore how broadly this method can be applied by examining its effects in reasoning with executable code and reasoning with common sense. We also explore how to apply this approach efficiently to extremely large language models using proxy-tuning. Experiment results on multilingual reasoning benchmarks mGSM, mSVAMP and xCSQA demonstrate that the question alignment approach can be used to boost multilingual performance across diverse reasoning scenarios, model families, and sizes. For instance, when applied to the LLaMA2 models, our method brings an average accuracy improvements of 12.2% on mGSM even with the 70B model. To understand the mechanism of its success, we analyze representation space, chain-of-thought and translation data scales, which reveals how question translation training strengthens language alignment within LLMs and shapes their working patterns.

Tabular reasoning involves interpreting unstructured queries against structured tables, requiring a synthesis of textual understanding and symbolic reasoning. Existing methods rely on either of the approaches and are constrained by their respective limitations. Textual reasoning excels in semantic interpretation unlike symbolic reasoning (SQL logic), but falls short in mathematical reasoning where SQL excels. In this paper, we introduce a novel algorithm H-STAR, comprising table extraction and adaptive reasoning, integrating both symbolic and semantic (text-based) approaches. To enhance evidence extraction, H-STAR employs a multi-view approach, incorporating step-by-step row and column retrieval. It also adapts reasoning strategies based on question types, utilizing symbolic reasoning for quantitative and logical tasks, and semantic reasoning for direct lookup and complex lexical queries. Our extensive experiments demonstrate that H-STAR significantly outperforms state-of-the-art methods across three tabular question-answering (QA) and fact-verification datasets, underscoring its effectiveness and efficiency.