Carl Zeiss Meditec AG
Abstract:Vision-language models (VLMs) have been applied to robot task planning problems, where the robot receives a task in natural language and generates plans based on visual inputs. While current VLMs have demonstrated strong vision-language understanding capabilities, their performance is still far from being satisfactory in planning tasks. At the same time, although classical task planners, such as PDDL-based, are strong in planning for long-horizon tasks, they do not work well in open worlds where unforeseen situations are common. In this paper, we propose a novel task planning and execution framework, called DKPROMPT, which automates VLM prompting using domain knowledge in PDDL for classical planning in open worlds. Results from quantitative experiments show that DKPROMPT outperforms classical planning, pure VLM-based and a few other competitive baselines in task completion rate.
Abstract:We introduce SpreadsheetBench, a challenging spreadsheet manipulation benchmark exclusively derived from real-world scenarios, designed to immerse current large language models (LLMs) in the actual workflow of spreadsheet users. Unlike existing benchmarks that rely on synthesized queries and simplified spreadsheet files, SpreadsheetBench is built from 912 real questions gathered from online Excel forums, which reflect the intricate needs of users. The associated spreadsheets from the forums contain a variety of tabular data such as multiple tables, non-standard relational tables, and abundant non-textual elements. Furthermore, we propose a more reliable evaluation metric akin to online judge platforms, where multiple spreadsheet files are created as test cases for each instruction, ensuring the evaluation of robust solutions capable of handling spreadsheets with varying values. Our comprehensive evaluation of various LLMs under both single-round and multi-round inference settings reveals a substantial gap between the state-of-the-art (SOTA) models and human performance, highlighting the benchmark's difficulty.
Abstract:We introduce ChatGLM, an evolving family of large language models that we have been developing over time. This report primarily focuses on the GLM-4 language series, which includes GLM-4, GLM-4-Air, and GLM-4-9B. They represent our most capable models that are trained with all the insights and lessons gained from the preceding three generations of ChatGLM. To date, the GLM-4 models are pre-trained on ten trillions of tokens mostly in Chinese and English, along with a small set of corpus from 24 languages, and aligned primarily for Chinese and English usage. The high-quality alignment is achieved via a multi-stage post-training process, which involves supervised fine-tuning and learning from human feedback. Evaluations show that GLM-4 1) closely rivals or outperforms GPT-4 in terms of general metrics such as MMLU, GSM8K, MATH, BBH, GPQA, and HumanEval, 2) gets close to GPT-4-Turbo in instruction following as measured by IFEval, 3) matches GPT-4 Turbo (128K) and Claude 3 for long context tasks, and 4) outperforms GPT-4 in Chinese alignments as measured by AlignBench. The GLM-4 All Tools model is further aligned to understand user intent and autonomously decide when and which tool(s) touse -- including web browser, Python interpreter, text-to-image model, and user-defined functions -- to effectively complete complex tasks. In practical applications, it matches and even surpasses GPT-4 All Tools in tasks like accessing online information via web browsing and solving math problems using Python interpreter. Over the course, we have open-sourced a series of models, including ChatGLM-6B (three generations), GLM-4-9B (128K, 1M), GLM-4V-9B, WebGLM, and CodeGeeX, attracting over 10 million downloads on Hugging face in the year 2023 alone. The open models can be accessed through https://github.com/THUDM and https://huggingface.co/THUDM.
Abstract:Evaluating the alignment capabilities of large Vision-Language Models (VLMs) is essential for determining their effectiveness as helpful assistants. However, existing benchmarks primarily focus on basic abilities using nonverbal methods, such as yes-no and multiple-choice questions. In this paper, we address this gap by introducing AlignMMBench, a comprehensive alignment benchmark specifically designed for emerging Chinese VLMs. This benchmark is meticulously curated from real-world scenarios and Chinese Internet sources, encompassing thirteen specific tasks across three categories, and includes both single-turn and multi-turn dialogue scenarios. Incorporating a prompt rewrite strategy, AlignMMBench encompasses 1,054 images and 4,978 question-answer pairs. To facilitate the evaluation pipeline, we propose CritiqueVLM, a rule-calibrated evaluator that exceeds GPT-4's evaluation ability. Finally, we report the performance of representative VLMs on AlignMMBench, offering insights into the capabilities and limitations of different VLM architectures. All evaluation codes and data are available on https://alignmmbench.github.io.
Abstract:Recent progress in multimodal large language models has markedly enhanced the understanding of short videos (typically under one minute), and several evaluation datasets have emerged accordingly. However, these advancements fall short of meeting the demands of real-world applications such as embodied intelligence for long-term decision-making, in-depth movie reviews and discussions, and live sports commentary, all of which require comprehension of long videos spanning several hours. To address this gap, we introduce LVBench, a benchmark specifically designed for long video understanding. Our dataset comprises publicly sourced videos and encompasses a diverse set of tasks aimed at long video comprehension and information extraction. LVBench is designed to challenge multimodal models to demonstrate long-term memory and extended comprehension capabilities. Our extensive evaluations reveal that current multimodal models still underperform on these demanding long video understanding tasks. Through LVBench, we aim to spur the development of more advanced models capable of tackling the complexities of long video comprehension. Our data and code are publicly available at: https://lvbench.github.io.
Abstract:Most recent multispectral object detectors employ a two-branch structure to extract features from RGB and thermal images. While the two-branch structure achieves better performance than a single-branch structure, it overlooks inference efficiency. This conflict is increasingly aggressive, as recent works solely pursue higher performance rather than both performance and efficiency. In this paper, we address this issue by improving the performance of efficient single-branch structures. We revisit the reasons causing the performance gap between these structures. For the first time, we reveal the information interference problem in the naive early-fusion strategy adopted by previous single-branch structures. Besides, we find that the domain gap between multispectral images, and weak feature representation of the single-branch structure are also key obstacles for performance. Focusing on these three problems, we propose corresponding solutions, including a novel shape-priority early-fusion strategy, a weakly supervised learning method, and a core knowledge distillation technique. Experiments demonstrate that single-branch networks equipped with these three contributions achieve significant performance enhancements while retaining high efficiency. Our code will be available at \url{https://github.com/XueZ-phd/Efficient-RGB-T-Early-Fusion-Detection}.
Abstract:Recent progress in large-scale scene rendering has yielded Neural Radiance Fields (NeRF)-based models with an impressive ability to synthesize scenes across small objects and indoor scenes. Nevertheless, extending this idea to large-scale aerial rendering poses two critical problems. Firstly, a single NeRF cannot render the entire scene with high-precision for complex large-scale aerial datasets since the sampling range along each view ray is insufficient to cover buildings adequately. Secondly, traditional NeRFs are infeasible to train on one GPU to enable interactive fly-throughs for modeling massive images. Instead, existing methods typically separate the whole scene into multiple regions and train a NeRF on each region, which are unaccustomed to different flight trajectories and difficult to achieve fast rendering. To that end, we propose Aerial-NeRF with three innovative modifications for jointly adapting NeRF in large-scale aerial rendering: (1) Designing an adaptive spatial partitioning and selection method based on drones' poses to adapt different flight trajectories; (2) Using similarity of poses instead of (expert) network for rendering speedup to determine which region a new viewpoint belongs to; (3) Developing an adaptive sampling approach for rendering performance improvement to cover the entire buildings at different heights. Extensive experiments have conducted to verify the effectiveness and efficiency of Aerial-NeRF, and new state-of-the-art results have been achieved on two public large-scale aerial datasets and presented SCUTic dataset. Note that our model allows us to perform rendering over 4 times as fast as compared to multiple competitors. Our dataset, code, and model are publicly available at https://drliuqi.github.io/.
Abstract:Large language models (LLMs) have manifested strong ability to generate codes for productive activities. However, current benchmarks for code synthesis, such as HumanEval, MBPP, and DS-1000, are predominantly oriented towards introductory tasks on algorithm and data science, insufficiently satisfying challenging requirements prevalent in real-world coding. To fill this gap, we propose NaturalCodeBench (NCB), a challenging code benchmark designed to mirror the complexity and variety of scenarios in real coding tasks. NCB comprises 402 high-quality problems in Python and Java, meticulously selected from natural user queries from online coding services, covering 6 different domains. Noting the extraordinary difficulty in creating testing cases for real-world queries, we also introduce a semi-automated pipeline to enhance the efficiency of test case construction. Comparing with manual solutions, it achieves an efficiency increase of more than 4 times. Our systematic experiments on 39 LLMs find that performance gaps on NCB between models with close HumanEval scores could still be significant, indicating a lack of focus on practical code synthesis scenarios or over-specified optimization on HumanEval. On the other hand, even the best-performing GPT-4 is still far from satisfying on NCB. The evaluation toolkit and development set are available at https://github.com/THUDM/NaturalCodeBench.
Abstract:Existing neural radiance fields (NeRF)-based novel view synthesis methods for large-scale outdoor scenes are mainly built on a single altitude. Moreover, they often require a priori camera shooting height and scene scope, leading to inefficient and impractical applications when camera altitude changes. In this work, we propose an end-to-end framework, termed AG-NeRF, and seek to reduce the training cost of building good reconstructions by synthesizing free-viewpoint images based on varying altitudes of scenes. Specifically, to tackle the detail variation problem from low altitude (drone-level) to high altitude (satellite-level), a source image selection method and an attention-based feature fusion approach are developed to extract and fuse the most relevant features of target view from multi-height images for high-fidelity rendering. Extensive experiments demonstrate that AG-NeRF achieves SOTA performance on 56 Leonard and Transamerica benchmarks and only requires a half hour of training time to reach the competitive PSNR as compared to the latest BungeeNeRF.
Abstract:Large language models (LLMs) have fueled many intelligent agent tasks, such as web navigation -- but most existing agents perform far from satisfying in real-world webpages due to three factors: (1) the versatility of actions on webpages, (2) HTML text exceeding model processing capacity, and (3) the complexity of decision-making due to the open-domain nature of web. In light of the challenge, we develop AutoWebGLM, a GPT-4-outperforming automated web navigation agent built upon ChatGLM3-6B. Inspired by human browsing patterns, we design an HTML simplification algorithm to represent webpages, preserving vital information succinctly. We employ a hybrid human-AI method to build web browsing data for curriculum training. Then, we bootstrap the model by reinforcement learning and rejection sampling to further facilitate webpage comprehension, browser operations, and efficient task decomposition by itself. For testing, we establish a bilingual benchmark -- AutoWebBench -- for real-world web browsing tasks. We evaluate AutoWebGLM across diverse web navigation benchmarks, revealing its improvements but also underlying challenges to tackle real environments. Related code, model, and data will be released at \url{https://github.com/THUDM/AutoWebGLM}.