Seemingly Plausible Distractors in Multi-Hop Reasoning: Are Large Language Models Attentive Readers?
State-of-the-art Large Language Models (LLMs) are accredited with an increasing number of different capabilities, ranging from reading comprehension, over advanced mathematical and reasoning skills to possessing scientific knowledge. In this paper we focus on their multi-hop reasoning capability: the ability to identify and integrate information from multiple textual sources. Given the concerns with the presence of simplifying cues in existing multi-hop reasoning benchmarks, which allow models to circumvent the reasoning requirement, we set out to investigate, whether LLMs are prone to exploiting such simplifying cues. We find evidence that they indeed circumvent the requirement to perform multi-hop reasoning, but they do so in more subtle ways than what was reported about their fine-tuned pre-trained language model (PLM) predecessors. Motivated by this finding, we propose a challenging multi-hop reasoning benchmark, by generating seemingly plausible multi-hop reasoning chains, which ultimately lead to incorrect answers. We evaluate multiple open and proprietary state-of-the-art LLMs, and find that their performance to perform multi-hop reasoning is affected, as indicated by up to 45% relative decrease in F1 score when presented with such seemingly plausible alternatives. We conduct a deeper analysis and find evidence that while LLMs tend to ignore misleading lexical cues, misleading reasoning paths indeed present a significant challenge.
相關內容
While Large Language Models (LLMs) have demonstrated remarkable performance in certain dimensions, their ability to express implicit language cues that human use for effective communication remains unclear. This paper presents ExpressivityArena, a Python library for measuring the implicit communication abilities of LLMs. We provide a comprehensive framework to evaluate expressivity of arbitrary LLMs and explore its practical implications. To this end, we refine the definition and measurements of ``expressivity,'' and use our framework in a set of small experiments. These experiments test LLMs in creative and logical tasks such as poetry, coding, and emotion-based responses. They are then evaluated by an automated grader, through ExpressivityArena, which we verify to be the most pragmatic for testing expressivity. Building on these experiments, we deepen our understanding of the expressivity of LLMs by assessing their ability to remain expressive in conversations. Our findings indicate that LLMs are capable of generating and understanding expressive content, however, with some limitations. These insights will inform the future development and deployment of expressive LLMs. We provide the code for ExpressivityArena alongside our paper.
The increase in parameter size of multimodal large language models (MLLMs) introduces significant capabilities, particularly in-context learning, where MLLMs enhance task performance without updating pre-trained parameters. This effectiveness, however, hinges on the appropriate selection of in-context examples, a process that is currently biased towards visual data, overlooking textual information. Furthermore, the area of supervised retrievers for MLLMs, crucial for optimal in-context example selection, continues to be uninvestigated. Our study offers an in-depth evaluation of the impact of textual information on the unsupervised selection of in-context examples in multimodal contexts, uncovering a notable sensitivity of retriever performance to the employed modalities. Responding to this, we introduce a novel supervised MLLM-retriever MSIER that employs a neural network to select examples that enhance multimodal in-context learning efficiency. This approach is validated through extensive testing across three distinct tasks, demonstrating the method's effectiveness. Additionally, we investigate the influence of modalities on our supervised retrieval method's training and pinpoint factors contributing to our model's success. This exploration paves the way for future advancements, highlighting the potential for refined in-context learning in MLLMs through the strategic use of multimodal data.
Transformer-based models have achieved remarkable success in various Natural Language Processing (NLP) tasks, yet their ability to handle long documents is constrained by computational limitations. Traditional approaches, such as truncating inputs, sparse self-attention, and chunking, attempt to mitigate these issues, but they often lead to information loss and hinder the model's ability to capture long-range dependencies. In this paper, we introduce ChuLo, a novel chunk representation method for long document classification that addresses these limitations. Our ChuLo groups input tokens using unsupervised keyphrase extraction, emphasizing semantically important keyphrase based chunk to retain core document content while reducing input length. This approach minimizes information loss and improves the efficiency of Transformer-based models. Preserving all tokens in long document understanding, especially token classification tasks, is especially important to ensure that fine-grained annotations, which depend on the entire sequence context, are not lost. We evaluate our method on multiple long document classification tasks and long document token classification tasks, demonstrating its effectiveness through comprehensive qualitative and quantitative analyses.
Test collections are information retrieval tools that allow researchers to quickly and easily evaluate ranking algorithms. While test collections have become an integral part of IR research, the process of data creation involves significant efforts in manual annotations, which often makes it very expensive and time-consuming. Thus, the test collections could become small when the budget is limited, which may lead to unstable evaluations. As an alternative, recent studies have proposed the use of large language models (LLMs) to completely replace human assessors. However, while LLMs seem to somewhat correlate with human judgments, they are not perfect and often show bias. Moreover, even if a well-performing LLM or prompt is found on one dataset, there is no guarantee that it will perform similarly in practice, due to difference in tasks and data. Thus a complete replacement with LLMs is argued to be too risky and not fully trustable. Thus, in this paper, we propose \textbf{L}LM-\textbf{A}ssisted \textbf{R}elevance \textbf{A}ssessments (\textbf{LARA}), an effective method to balance manual annotations with LLM annotations, which helps to make a rich and reliable test collection. We use the LLM's predicted relevance probabilities in order to select the most profitable documents to manually annotate under a budget constraint. While solely relying on LLM's predicted probabilities to manually annotate performs fairly well, with theoretical reasoning, LARA guides the human annotation process even more effectively via online calibration learning. Then, using the calibration model learned from the limited manual annotations, LARA debiases the LLM predictions to annotate the remaining non-assessed data. Empirical evaluations on TREC-COVID and TREC-8 Ad Hoc datasets show that LARA outperforms the alternative solutions under almost any budget constraint.
In recent years, Large Language Models (LLMs) have demonstrated remarkable versatility across various applications, including natural language understanding, domain-specific knowledge tasks, etc. However, applying LLMs to complex, high-stakes domains like finance requires rigorous evaluation to ensure reliability, accuracy, and compliance with industry standards. To address this need, we conduct a comprehensive and comparative study on three state-of-the-art LLMs, GLM-4, Mistral-NeMo, and LLaMA3.1, focusing on their effectiveness in generating automated financial reports. Our primary motivation is to explore how these models can be harnessed within finance, a field demanding precision, contextual relevance, and robustness against erroneous or misleading information. By examining each model's capabilities, we aim to provide an insightful assessment of their strengths and limitations. Our paper offers benchmarks for financial report analysis, encompassing proposed metrics such as ROUGE-1, BERT Score, and LLM Score. We introduce an innovative evaluation framework that integrates both quantitative metrics (e.g., precision, recall) and qualitative analyses (e.g., contextual fit, consistency) to provide a holistic view of each model's output quality. Additionally, we make our financial dataset publicly available, inviting researchers and practitioners to leverage, scrutinize, and enhance our findings through broader community engagement and collaborative improvement. Our dataset is available on huggingface.
Detecting evidence within the context is a key step in the process of reasoning task. Evaluating and enhancing the capabilities of LLMs in evidence detection will strengthen context-based reasoning performance. This paper proposes a benchmark called DetectBench for verifying the ability to detect and piece together implicit evidence within a long context. DetectBench contains 3,928 multiple-choice questions, with an average of 994 tokens per question. Each question contains an average of 4.55 pieces of implicit evidence, and solving the problem typically requires 7.62 logical jumps to find the correct answer. To enhance the performance of LLMs in evidence detection, this paper proposes Detective Reasoning Prompt and Finetune. Experiments demonstrate that the existing LLMs' abilities to detect evidence in long contexts are far inferior to humans. However, the Detective Reasoning Prompt effectively enhances the capability of powerful LLMs in evidence detection, while the Finetuning method shows significant effects in enhancing the performance of weaker LLMs. Moreover, when the abilities of LLMs in evidence detection are improved, their final reasoning performance is also enhanced accordingly.
Large Language Models for Code (LLMs4Code) have been found to exhibit outstanding performance in the software engineering domain, especially the remarkable performance in coding tasks. However, even the most advanced LLMs4Code can inevitably contain incorrect or outdated code knowledge. Due to the high cost of training LLMs4Code, it is impractical to re-train the models for fixing these problematic code knowledge. Model editing is a new technical field for effectively and efficiently correcting erroneous knowledge in LLMs, where various model editing techniques and benchmarks have been proposed recently. Despite that, a comprehensive study that thoroughly compares and analyzes the performance of the state-of-the-art model editing techniques for adapting the knowledge within LLMs4Code across various code-related tasks is notably absent. To bridge this gap, we perform the first systematic study on applying state-of-the-art model editing approaches to repair the inaccuracy of LLMs4Code. To that end, we introduce a benchmark named CLMEEval, which consists of two datasets, i.e., CoNaLa-Edit (CNLE) with 21K+ code generation samples and CodeSearchNet-Edit (CSNE) with 16K+ code summarization samples. With the help of CLMEEval, we evaluate six advanced model editing techniques on three LLMs4Code: CodeLlama (7B), CodeQwen1.5 (7B), and Stable-Code (3B). Our findings include that the external memorization-based GRACE approach achieves the best knowledge editing effectiveness and specificity (the editing does not influence untargeted knowledge), while generalization (whether the editing can generalize to other semantically-identical inputs) is a universal challenge for existing techniques. Furthermore, building on in-depth case analysis, we introduce an enhanced version of GRACE called A-GRACE, which incorporates contrastive learning to better capture the semantics of the inputs.
Recurrent Neural Networks (RNNs) are widely recognized for their proficiency in modeling temporal dependencies, making them highly prevalent in sequential data processing applications. Nevertheless, vanilla RNNs are confronted with the well-known issue of gradient vanishing and exploding, posing a significant challenge for learning and establishing long-range dependencies. Additionally, gated RNNs tend to be over-parameterized, resulting in poor computational efficiency and network generalization. To address these challenges, this paper proposes a novel Delayed Memory Unit (DMU). The DMU incorporates a delay line structure along with delay gates into vanilla RNN, thereby enhancing temporal interaction and facilitating temporal credit assignment. Specifically, the DMU is designed to directly distribute the input information to the optimal time instant in the future, rather than aggregating and redistributing it over time through intricate network dynamics. Our proposed DMU demonstrates superior temporal modeling capabilities across a broad range of sequential modeling tasks, utilizing considerably fewer parameters than other state-of-the-art gated RNN models in applications such as speech recognition, radar gesture recognition, ECG waveform segmentation, and permuted sequential image classification.
Compared with cheap addition operation, multiplication operation is of much higher computation complexity. The widely-used convolutions in deep neural networks are exactly cross-correlation to measure the similarity between input feature and convolution filters, which involves massive multiplications between float values. In this paper, we present adder networks (AdderNets) to trade these massive multiplications in deep neural networks, especially convolutional neural networks (CNNs), for much cheaper additions to reduce computation costs. In AdderNets, we take the $\ell_1$-norm distance between filters and input feature as the output response. The influence of this new similarity measure on the optimization of neural network have been thoroughly analyzed. To achieve a better performance, we develop a special back-propagation approach for AdderNets by investigating the full-precision gradient. We then propose an adaptive learning rate strategy to enhance the training procedure of AdderNets according to the magnitude of each neuron's gradient. As a result, the proposed AdderNets can achieve 74.9% Top-1 accuracy 91.7% Top-5 accuracy using ResNet-50 on the ImageNet dataset without any multiplication in convolution layer.
The problem of Multiple Object Tracking (MOT) consists in following the trajectory of different objects in a sequence, usually a video. In recent years, with the rise of Deep Learning, the algorithms that provide a solution to this problem have benefited from the representational power of deep models. This paper provides a comprehensive survey on works that employ Deep Learning models to solve the task of MOT on single-camera videos. Four main steps in MOT algorithms are identified, and an in-depth review of how Deep Learning was employed in each one of these stages is presented. A complete experimental comparison of the presented works on the three MOTChallenge datasets is also provided, identifying a number of similarities among the top-performing methods and presenting some possible future research directions.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191