DOCBENCH: A Benchmark for Evaluating LLM-based Document Reading Systems

Recently, there has been a growing interest among large language model (LLM) developers in LLM-based document reading systems, which enable users to upload their own documents and pose questions related to the document contents, going beyond simple reading comprehension tasks. Consequently, these systems have been carefully designed to tackle challenges such as file parsing, metadata extraction, multi-modal information understanding and long-context reading. However, no current benchmark exists to evaluate their performance in such scenarios, where a raw file and questions are provided as input, and a corresponding response is expected as output. In this paper, we introduce DocBench, a new benchmark designed to evaluate LLM-based document reading systems. Our benchmark involves a meticulously crafted process, including the recruitment of human annotators and the generation of synthetic questions. It includes 229 real documents and 1,102 questions, spanning across five different domains and four major types of questions. We evaluate both proprietary LLM-based systems accessible via web interfaces or APIs, and a parse-then-read pipeline employing open-source LLMs. Our evaluations reveal noticeable gaps between existing LLM-based document reading systems and human performance, underscoring the challenges of developing proficient systems. To summarize, DocBench aims to establish a standardized benchmark for evaluating LLM-based document reading systems under diverse real-world scenarios, thereby guiding future advancements in this research area.

TASeg: Temporal Aggregation Network for LiDAR Semantic Segmentation

Training deep models for LiDAR semantic segmentation is challenging due to the inherent sparsity of point clouds. Utilizing temporal data is a natural remedy against the sparsity problem as it makes the input signal denser. However, previous multi-frame fusion algorithms fall short in utilizing sufficient temporal information due to the memory constraint, and they also ignore the informative temporal images. To fully exploit rich information hidden in long-term temporal point clouds and images, we present the Temporal Aggregation Network, termed TASeg. Specifically, we propose a Temporal LiDAR Aggregation and Distillation (TLAD) algorithm, which leverages historical priors to assign different aggregation steps for different classes. It can largely reduce memory and time overhead while achieving higher accuracy. Besides, TLAD trains a teacher injected with gt priors to distill the model, further boosting the performance. To make full use of temporal images, we design a Temporal Image Aggregation and Fusion (TIAF) module, which can greatly expand the camera FOV and enhance the present features. Temporal LiDAR points in the camera FOV are used as mediums to transform temporal image features to the present coordinate for temporal multi-modal fusion. Moreover, we develop a Static-Moving Switch Augmentation (SMSA) algorithm, which utilizes sufficient temporal information to enable objects to switch their motion states freely, thus greatly increasing static and moving training samples. Our TASeg ranks 1st on three challenging tracks, i.e., SemanticKITTI single-scan track, multi-scan track and nuScenes LiDAR segmentation track, strongly demonstrating the superiority of our method. Codes are available at https://github.com/LittlePey/TASeg.

GLBench: A Comprehensive Benchmark for Graph with Large Language Models

The emergence of large language models (LLMs) has revolutionized the way we interact with graphs, leading to a new paradigm called GraphLLM. Despite the rapid development of GraphLLM methods in recent years, the progress and understanding of this field remain unclear due to the lack of a benchmark with consistent experimental protocols. To bridge this gap, we introduce GLBench, the first comprehensive benchmark for evaluating GraphLLM methods in both supervised and zero-shot scenarios. GLBench provides a fair and thorough evaluation of different categories of GraphLLM methods, along with traditional baselines such as graph neural networks. Through extensive experiments on a collection of real-world datasets with consistent data processing and splitting strategies, we have uncovered several key findings. Firstly, GraphLLM methods outperform traditional baselines in supervised settings, with LLM-as-enhancers showing the most robust performance. However, using LLMs as predictors is less effective and often leads to uncontrollable output issues. We also notice that no clear scaling laws exist for current GraphLLM methods. In addition, both structures and semantics are crucial for effective zero-shot transfer, and our proposed simple baseline can even outperform several models tailored for zero-shot scenarios. The data and code of the benchmark can be found at https://github.com/NineAbyss/GLBench.

Not All Preference Pairs Are Created Equal: A Recipe for Annotation-Efficient Iterative Preference Learning

Iterative preference learning, though yielding superior performances, requires online annotated preference labels. In this work, we study strategies to select worth-annotating response pairs for cost-efficient annotation while achieving competitive or even better performances compared with the random selection baseline for iterative preference learning. Built on assumptions regarding uncertainty and distribution shifts, we propose a comparative view to rank the implicit reward margins as predicted by DPO to select the response pairs that yield more benefits. Through extensive experiments, we show that annotating those response pairs with small margins is generally better than large or random, under both single- and multi-iteration scenarios. Besides, our empirical results suggest allocating more annotation budgets in the earlier iterations rather than later across multiple iterations.

On the Transformations across Reward Model, Parameter Update, and In-Context Prompt

Despite the general capabilities of pre-trained large language models (LLMs), they still need further adaptation to better serve practical applications. In this paper, we demonstrate the interchangeability of three popular and distinct adaptation tools: parameter updating, reward modeling, and in-context prompting. This interchangeability establishes a triangular framework with six transformation directions, each of which facilitates a variety of applications. Our work offers a holistic view that unifies numerous existing studies and suggests potential research directions. We envision our work as a useful roadmap for future research on LLMs.

ChartMimic: Evaluating LMM's Cross-Modal Reasoning Capability via Chart-to-Code Generation

We introduce a new benchmark, ChartMimic, aimed at assessing the visually-grounded code generation capabilities of large multimodal models (LMMs). ChartMimic utilizes information-intensive visual charts and textual instructions as inputs, requiring LMMs to generate the corresponding code for chart rendering. ChartMimic includes 1,000 human-curated (figure, instruction, code) triplets, which represent the authentic chart use cases found in scientific papers across various domains(e.g., Physics, Computer Science, Economics, etc). These charts span 18 regular types and 4 advanced types, diversifying into 191 subcategories. Furthermore, we propose multi-level evaluation metrics to provide an automatic and thorough assessment of the output code and the rendered charts. Unlike existing code generation benchmarks, ChartMimic places emphasis on evaluating LMMs' capacity to harmonize a blend of cognitive capabilities, encompassing visual understanding, code generation, and cross-modal reasoning. The evaluation of 3 proprietary models and 11 open-weight models highlights the substantial challenges posed by ChartMimic. Even the advanced GPT-4V, Claude-3-opus only achieve an average score of 73.2 and 53.7, respectively, indicating significant room for improvement. We anticipate that ChartMimic will inspire the development of LMMs, advancing the pursuit of artificial general intelligence.

On the Worst Prompt Performance of Large Language Models

The performance of large language models (LLMs) is acutely sensitive to the phrasing of prompts, which raises significant concerns about their reliability in real-world scenarios. Existing studies often divide prompts into task-level instructions and case-level inputs and primarily focus on evaluating and improving robustness against variations in tasks-level instructions. However, this setup fails to fully address the diversity of real-world user queries and assumes the existence of task-specific datasets. To address these limitations, we introduce RobustAlpacaEval, a new benchmark that consists of semantically equivalent case-level queries and emphasizes the importance of using the worst prompt performance to gauge the lower bound of model performance. Extensive experiments on RobustAlpacaEval with ChatGPT and six open-source LLMs from the Llama, Mistral, and Gemma families uncover substantial variability in model performance; for instance, a difference of 45.48% between the worst and best performance for the Llama-2-70B-chat model, with its worst performance dipping as low as 9.38%. We further illustrate the difficulty in identifying the worst prompt from both model-agnostic and model-dependent perspectives, emphasizing the absence of a shortcut to characterize the worst prompt. We also attempt to enhance the worst prompt performance using existing prompt engineering and prompt consistency methods, but find that their impact is limited. These findings underscore the need to create more resilient LLMs that can maintain high performance across diverse prompts.

Unchosen Experts Can Contribute Too: Unleashing MoE Models' Power by Self-Contrast

Mixture-of-Experts (MoE) has emerged as a prominent architecture for scaling model size while maintaining computational efficiency. In MoE, each token in the input sequence activates a different subset of experts determined by a routing mechanism. However, the unchosen experts in MoE models do not contribute to the output, potentially leading to underutilization of the model's capacity. In this work, we first conduct exploratory studies to demonstrate that increasing the number of activated experts does not necessarily improve and can even degrade the output quality. Then, we show that output distributions from an MoE model using different routing strategies substantially differ, indicating that different experts do not always act synergistically. Motivated by these findings, we propose Self-Contrast Mixture-of-Experts (SCMoE), a training-free strategy that utilizes unchosen experts in a self-contrast manner during inference. In SCMoE, the next-token probabilities are determined by contrasting the outputs from strong and weak activation using the same MoE model. Our method is conceptually simple and computationally lightweight, as it incurs minimal latency compared to greedy decoding. Experiments on several benchmarks (GSM8K, StrategyQA, MBPP and HumanEval) demonstrate that SCMoE can consistently enhance Mixtral 8x7B's reasoning capability across various domains. For example, it improves the accuracy on GSM8K from 61.79 to 66.94. Moreover, combining SCMoE with self-consistency yields additional gains, increasing major@20 accuracy from 75.59 to 78.31.

Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction

Supervised fine-tuning (SFT) on instruction-following corpus is a crucial approach toward the alignment of large language models (LLMs). However, the performance of LLMs on standard knowledge and reasoning benchmarks tends to suffer from deterioration at the latter stage of the SFT process, echoing the phenomenon of alignment tax. Through our pilot study, we put a hypothesis that the data biases are probably one cause behind the phenomenon. To address the issue, we introduce a simple disperse-then-merge framework. To be concrete, we disperse the instruction-following data into portions and train multiple sub-models using different data portions. Then we merge multiple models into a single one via model merging techniques. Despite its simplicity, our framework outperforms various sophisticated methods such as data curation and training regularization on a series of standard knowledge and reasoning benchmarks.

G2LTraj: A Global-to-Local Generation Approach for Trajectory Prediction

Predicting future trajectories of traffic agents accurately holds substantial importance in various applications such as autonomous driving. Previous methods commonly infer all future steps of an agent either recursively or simultaneously. However, the recursive strategy suffers from the accumulated error, while the simultaneous strategy overlooks the constraints among future steps, resulting in kinematically infeasible predictions. To address these issues, in this paper, we propose G2LTraj, a plug-and-play global-to-local generation approach for trajectory prediction. Specifically, we generate a series of global key steps that uniformly cover the entire future time range. Subsequently, the local intermediate steps between the adjacent key steps are recursively filled in. In this way, we prevent the accumulated error from propagating beyond the adjacent key steps. Moreover, to boost the kinematical feasibility, we not only introduce the spatial constraints among key steps but also strengthen the temporal constraints among the intermediate steps. Finally, to ensure the optimal granularity of key steps, we design a selectable granularity strategy that caters to each predicted trajectory. Our G2LTraj significantly improves the performance of seven existing trajectory predictors across the ETH, UCY and nuScenes datasets. Experimental results demonstrate its effectiveness. Code will be available at https://github.com/Zhanwei-Z/G2LTraj.