Siddharth Jha

Lutfi Eren Erdogan, Nicholas Lee, Siddharth Jha et al.

Recent large language models (LLMs) have enabled the development of advanced agentic systems that can integrate various tools and APIs to fulfill user queries through function calling. However, the deployment of these LLMs on the edge has not been explored since they typically require cloud-based infrastructure due to their substantial model size and computational demands. To this end, we present TinyAgent, an end-to-end framework for training and deploying task-specific small language model agents capable of function calling for driving agentic systems at the edge. We first show how to enable accurate function calling for open-source models via the LLMCompiler framework. We then systematically curate a high-quality dataset for function calling, which we use to fine-tune two small language models, TinyAgent-1.1B and 7B. For efficient inference, we introduce a novel tool retrieval method to reduce the input prompt length and utilize quantization to further accelerate the inference speed. As a driving application, we demonstrate a local Siri-like system for Apple's MacBook that can execute user commands through text or voice input. Our results show that our models can achieve, and even surpass, the function-calling capabilities of larger models like GPT-4-Turbo, while being fully deployed at the edge. We open-source our dataset, models, and installable package and provide a demo video for our MacBook assistant agent.

Suhong Moon, Siddharth Jha, Lutfi Eren Erdogan et al.

Recent advancements in function calling and tool use have significantly enhanced the capabilities of large language models (LLMs) by enabling them to interact with external information sources and execute complex tasks. However, the limited context window of LLMs presents challenges when a large number of tools are available, necessitating efficient methods to manage prompt length and maintain accuracy. Existing approaches, such as fine-tuning LLMs or leveraging their reasoning capabilities, either require frequent retraining or incur significant latency overhead. A more efficient solution involves training smaller models to retrieve the most relevant tools for a given query, although this requires high quality, domain-specific data. To address those challenges, we present a novel framework for generating synthetic data for tool retrieval applications and an efficient data-driven tool retrieval strategy using small encoder models. Empowered by LLMs, we create ToolBank, a new tool retrieval dataset that reflects real human user usages. For tool retrieval methodologies, we propose novel approaches: (1) Tool2Vec: usage-driven tool embedding generation for tool retrieval, (2) ToolRefiner: a staged retrieval method that iteratively improves the quality of retrieved tools, and (3) MLC: framing tool retrieval as a multi-label classification problem. With these new methods, we achieve improvements of up to 27.28 in Recall@K on the ToolBench dataset and 30.5 in Recall@K on ToolBank. Additionally, we present further experimental results to rigorously validate our methods. Our code is available at \url{https://github.com/SqueezeAILab/Tool2Vec}

Liana Patel, Siddharth Jha, Melissa Pan et al.

The semantic capabilities of large language models (LLMs) have the potential to enable rich analytics and reasoning over vast knowledge corpora. Unfortunately, existing systems either empirically optimize expensive LLM-powered operations with no performance guarantees, or serve a limited set of row-wise LLM operations, providing limited robustness, expressiveness and usability. We introduce semantic operators, the first formalism for declarative and general-purpose AI-based transformations based on natural language specifications (e.g., filtering, sorting, joining or aggregating records using natural language criteria). Each operator opens a rich space for execution plans, similar to relational operators. Our model specifies the expected behavior of each operator with a high-quality gold algorithm, and we develop an optimization framework that reduces cost, while providing accuracy guarantees with respect to a gold algorithm. Using this approach, we propose several novel optimizations to accelerate semantic filtering, joining, group-by and top-k operations by up to $1,000\times$. We implement semantic operators in the LOTUS system and demonstrate LOTUS' effectiveness on real, bulk-semantic processing applications, including fact-checking, biomedical multi-label classification, search, and topic analysis. We show that the semantic operator model is expressive, capturing state-of-the-art AI pipelines in a few operator calls, and making it easy to express new pipelines that match or exceed quality of recent LLM-based analytic systems by up to $170\%$, while offering accuracy guarantees. Overall, LOTUS programs match or exceed the accuracy of state-of-the-art AI pipelines for each task while running up to $3.6\times$ faster than the highest-quality baselines. LOTUS is publicly available at https://github.com/lotus-data/lotus.

Siddharth Jha, Lutfi Eren Erdogan, Sehoon Kim et al.

Long context inference presents challenges at the system level with increased compute and memory requirements, as well as from an accuracy perspective in being able to reason over long contexts. Recently, several methods have been proposed to compress the prompt to reduce the context length. However, there has been little work on comparing the different proposed methods across different tasks through a standardized analysis. This has led to conflicting results. To address this, here we perform a comprehensive characterization and evaluation of different prompt compression methods. In particular, we analyze extractive compression, summarization-based abstractive compression, and token pruning methods. Surprisingly, we find that extractive compression often outperforms all the other approaches, and enables up to 10x compression with minimal accuracy degradation. Interestingly, we also find that despite several recent claims, token pruning methods often lag behind extractive compression. We only found marginal improvements on summarization tasks.

Xiaoxuan Liu, Siddharth Jha, Alvin Cheung

As models continue to grow in size, the development of memory optimization methods (MOMs) has emerged as a solution to address the memory bottleneck encountered when training large models. To comprehensively examine the practical value of various MOMs, we have conducted a thorough analysis of existing literature from a systems perspective. Our analysis has revealed a notable challenge within the research community: the absence of standardized metrics for effectively evaluating the efficacy of MOMs. The scarcity of informative evaluation metrics hinders the ability of researchers and practitioners to compare and benchmark different approaches reliably. Consequently, drawing definitive conclusions and making informed decisions regarding the selection and application of MOMs becomes a challenging endeavor. To address the challenge, this paper summarizes the scenarios in which MOMs prove advantageous for model training. We propose the use of distinct evaluation metrics under different scenarios. By employing these metrics, we evaluate the prevailing MOMs and find that their benefits are not universal. We present insights derived from experiments and discuss the circumstances in which they can be advantageous.

Zhipeng Chang, Siddharth Jha, Yunfei Xia

The rapid development of remote sensing technologies have gained significant attention due to their ability to accurately localize, classify, and segment objects from aerial images. These technologies are commonly used in unmanned aerial vehicles (UAVs) equipped with high-resolution cameras or sensors to capture data over large areas. This data is useful for various applications, such as monitoring and inspecting cities, towns, and terrains. In this paper, we presented a method for classifying and segmenting city road traffic dashed lines from aerial images using deep learning models such as U-Net and SegNet. The annotated data is used to train these models, which are then used to classify and segment the aerial image into two classes: dashed lines and non-dashed lines. However, the deep learning model may not be able to identify all dashed lines due to poor painting or occlusion by trees or shadows. To address this issue, we proposed a method to add missed lines to the segmentation output. We also extracted the x and y coordinates of each dashed line from the segmentation output, which can be used by city planners to construct a CAD file for digital visualization of the roads.

Siddharth Jha, Coleman Hooper, Xiaoxuan Liu et al.

Many applications must provide low-latency LLM service to users or risk unacceptable user experience. However, over-provisioning resources to serve fluctuating request patterns is often prohibitively expensive. In this work, we present a best-effort serving system that employs deep reinforcement learning to adjust service quality based on the task distribution and system load. Our best-effort system can maintain availability with over 10x higher client request rates, serves above 96% of peak performance 4.1x more often, and serves above 98% of peak performance 2.3x more often than static serving on unpredictable workloads. Our learned router is robust to shifts in both the arrival and task distribution. Compared to static serving, learned best-effort serving allows for cost-efficient serving through increased hardware utility. Additionally, we argue that learned best-effort LLM serving is applicable in wide variety of settings and provides application developers great flexibility to meet their specific needs.

Het Shah, Siddhant Haldar, Rohit Ner et al.

This paper describes the design and development of a non-intrusive inertial speed sensor that can be reliably used to replace a conventional optical or hall effect-based speedometer on any kind of ground vehicle. The design allows for simple assembly-disassembly from tyre rims. The sensor design and data flow are explained. Algorithms and filters for pre-processing and processing the data are detailed. Comparison with a real optical encoder proves the accuracy of the proposed sensor. Finally, it is shown that factor graph-based localization is possible with the developed sensor.

Siddharth Jha, Himanshu Chaudhary, Swapnil Satardey et al.

In this paper, we describe the mechanical design, system overview, integration and control techniques associated with SKALA, a unique large-sized robot for carrying a person with physical disabilities, up and down staircases. As a regular wheelchair is unable to perform such a maneuver, the system functions as a non-conventional wheelchair with several intelligent features. We describe the unique mechanical design and the design choices associated with it. We showcase the embedded control architecture that allows for several different modes of teleoperation, all of which have been described in detail. We further investigate the architecture associated with the autonomous operation of the system.

Ayush Pandey, Siddharth Jha, Debashish Chakravarty

Modeling and control strategies for a design of an autonomous three wheeled mobile robot with front wheel steer is presented. Although, the three-wheel vehicle design with front wheel steer is common in automotive vehicles used often in public transport, but its advantages in navigation and localization of autonomous vehicles is seldom utilized. We present the system model for such a robotic vehicle. A PID controller for speed control is designed for the model obtained and has been implemented in a digital control framework. The trajectory control framework, which is a challenging task for such a three-wheeled robot has also been presented in the paper. The derived system model has been verified using experimental results obtained for the robot vehicle design. Controller performance and robustness issues have also been discussed briefly.

Ayush Pandey, Subhamoy Mahajan, Adarsh Kosta et al.

On the lines of the huge and varied efforts in the field of automation with respect to technology development and innovation of vehicles to make them run autonomously, this paper presents an innovation to a bicycle. A normal daily use bicycle was modified at low cost such that it runs autonomously, while maintaining its original form i.e. the manual drive. Hence, a bicycle which could be normally driven by any human and with a press of switch could run autonomously according to the needs of the user has been developed.