This study introduces a novel software tool leveraging large language model (LLM) prompts, designed to automate the generation of academic articles from Python code a significant advancement in the fields of biomedical informatics and computer science. Selected for its widespread adoption and analytical versatility, Python served as a foundational proof of concept; however, the underlying methodology and framework exhibit adaptability across various GitHub repo's underlining the tool's broad applicability (Harper 2024). By mitigating the traditionally time-intensive academic writing process, particularly in synthesizing complex datasets and coding outputs, this approach signifies a monumental leap towards streamlining research dissemination. The development was achieved without reliance on advanced language model agents, ensuring high fidelity in the automated generation of coherent and comprehensive academic content. This exploration not only validates the successful application and efficiency of the software but also projects how future integration of LLM agents which could amplify its capabilities, propelling towards a future where scientific findings are disseminated more swiftly and accessibly.
In patent prosecution, image-based retrieval systems for identifying similarities between current patent images and prior art are pivotal to ensure the novelty and non-obviousness of patent applications. Despite their growing popularity in recent years, existing attempts, while effective at recognizing images within the same patent, fail to deliver practical value due to their limited generalizability in retrieving relevant prior art. Moreover, this task inherently involves the challenges posed by the abstract visual features of patent images, the skewed distribution of image classifications, and the semantic information of image descriptions. Therefore, we propose a language-informed, distribution-aware multimodal approach to patent image feature learning, which enriches the semantic understanding of patent image by integrating Large Language Models and improves the performance of underrepresented classes with our proposed distribution-aware contrastive losses. Extensive experiments on DeepPatent2 dataset show that our proposed method achieves state-of-the-art or comparable performance in image-based patent retrieval with mAP +53.3%, Recall@10 +41.8%, and MRR@10 +51.9%. Furthermore, through an in-depth user analysis, we explore our model in aiding patent professionals in their image retrieval efforts, highlighting the model's real-world applicability and effectiveness.
In recent advances in automatic text recognition (ATR), deep neural networks have demonstrated the ability to implicitly capture language statistics, potentially reducing the need for traditional language models. This study directly addresses whether explicit language models, specifically n-gram models, still contribute to the performance of state-of-the-art deep learning architectures in the field of handwriting recognition. We evaluate two prominent neural network architectures, PyLaia and DAN, with and without the integration of explicit n-gram language models. Our experiments on three datasets - IAM, RIMES, and NorHand v2 - at both line and page level, investigate optimal parameters for n-gram models, including their order, weight, smoothing methods and tokenization level. The results show that incorporating character or subword n-gram models significantly improves the performance of ATR models on all datasets, challenging the notion that deep learning models alone are sufficient for optimal performance. In particular, the combination of DAN with a character language model outperforms current benchmarks, confirming the value of hybrid approaches in modern document analysis systems.
Large vision-language models have impressively promote the performance of 2D visual recognition under zero/few-shot scenarios. In this paper, we focus on exploiting the large vision-language model, i.e., CLIP, to address zero/few-shot 3D shape recognition based on multi-view representations. The key challenge for both tasks is to generate a discriminative descriptor of the 3D shape represented by multiple view images under the scenarios of either without explicit training (zero-shot 3D shape recognition) or training with a limited number of data (few-shot 3D shape recognition). We analyze that both tasks are relevant and can be considered simultaneously. Specifically, leveraging the descriptor which is effective for zero-shot inference to guide the tuning of the aggregated descriptor under the few-shot training can significantly improve the few-shot learning efficacy. Hence, we propose Prompt-Enhanced View Aggregation Network (PEVA-Net) to simultaneously address zero/few-shot 3D shape recognition. Under the zero-shot scenario, we propose to leverage the prompts built up from candidate categories to enhance the aggregation process of multiple view-associated visual features. The resulting aggregated feature serves for effective zero-shot recognition of the 3D shapes. Under the few-shot scenario, we first exploit a transformer encoder to aggregate the view-associated visual features into a global descriptor. To tune the encoder, together with the main classification loss, we propose a self-distillation scheme via a feature distillation loss by treating the zero-shot descriptor as the guidance signal for the few-shot descriptor. This scheme can significantly enhance the few-shot learning efficacy.
Existing automatic captioning methods for visual content face challenges such as lack of detail, content hallucination, and poor instruction following. In this work, we propose VisualFactChecker (VFC), a flexible training-free pipeline that generates high-fidelity and detailed captions for both 2D images and 3D objects. VFC consists of three steps: 1) proposal, where image-to-text captioning models propose multiple initial captions; 2) verification, where a large language model (LLM) utilizes tools such as object detection and VQA models to fact-check proposed captions; 3) captioning, where an LLM generates the final caption by summarizing caption proposals and the fact check verification results. In this step, VFC can flexibly generate captions in various styles following complex instructions. We conduct comprehensive captioning evaluations using four metrics: 1) CLIP-Score for image-text similarity; 2) CLIP-Image-Score for measuring the image-image similarity between the original and the reconstructed image generated by a text-to-image model using the caption. 3) human study on Amazon Mechanical Turk; 4) GPT-4V for fine-grained evaluation. Evaluation results show that VFC outperforms state-of-the-art open-sourced captioning methods for 2D images on the COCO dataset and 3D assets on the Objaverse dataset. Our study demonstrates that by combining open-source models into a pipeline, we can attain captioning capability comparable to proprietary models such as GPT-4V, despite being over 10x smaller in model size.
Data protection and privacy is becoming increasingly crucial in the digital era. Numerous companies depend on third-party vendors and service providers to carry out critical functions within their operations, encompassing tasks such as data handling and storage. However, this reliance introduces potential vulnerabilities, as these vendors' security measures and practices may not always align with the standards expected by regulatory bodies. Businesses are required, often under the penalty of law, to ensure compliance with the evolving regulatory rules. Interpreting and implementing these regulations pose challenges due to their complexity. Regulatory documents are extensive, demanding significant effort for interpretation, while vendor-drafted privacy policies often lack the detail required for full legal compliance, leading to ambiguity. To ensure a concise interpretation of the regulatory requirements and compliance of organizational privacy policy with said regulations, we propose a Large Language Model (LLM) and Semantic Web based approach for privacy compliance. In this paper, we develop the novel Privacy Policy Compliance Verification Knowledge Graph, PrivComp-KG. It is designed to efficiently store and retrieve comprehensive information concerning privacy policies, regulatory frameworks, and domain-specific knowledge pertaining to the legal landscape of privacy. Using Retrieval Augmented Generation, we identify the relevant sections in a privacy policy with corresponding regulatory rules. This information about individual privacy policies is populated into the PrivComp-KG. Combining this with the domain context and rules, the PrivComp-KG can be queried to check for compliance with privacy policies by each vendor against relevant policy regulations. We demonstrate the relevance of the PrivComp-KG, by verifying compliance of privacy policy documents for various organizations.
Out-of-distribution (OOD) problems in few-shot classification (FSC) occur when novel classes sampled from testing distributions differ from base classes drawn from training distributions, which considerably degrades the performance of deep learning models deployed in real-world applications. Recent studies suggest that the OOD problems in FSC mainly including: (a) cross-domain few-shot classification (CD-FSC) and (b) spurious-correlation few-shot classification (SC-FSC). Specifically, CD-FSC occurs when a classifier learns transferring knowledge from base classes drawn from seen training distributions but recognizes novel classes sampled from unseen testing distributions. In contrast, SC-FSC arises when a classifier relies on non-causal features (or contexts) that happen to be correlated with the labels (or concepts) in base classes but such relationships no longer hold during the model deployment. Despite CD-FSC has been extensively studied, SC-FSC remains understudied due to lack of the corresponding evaluation benchmarks. To this end, we present Meta Concept Context (MetaCoCo), a benchmark with spurious-correlation shifts collected from real-world scenarios. Moreover, to quantify the extent of spurious-correlation shifts of the presented MetaCoCo, we further propose a metric by using CLIP as a pre-trained vision-language model. Extensive experiments on the proposed benchmark are performed to evaluate the state-of-the-art methods in FSC, cross-domain shifts, and self-supervised learning. The experimental results show that the performance of the existing methods degrades significantly in the presence of spurious-correlation shifts. We open-source all codes of our benchmark and hope that the proposed MetaCoCo can facilitate future research on spurious-correlation shifts problems in FSC. The code is available at: https://github.com/remiMZ/MetaCoCo-ICLR24.
Large Language Models (LLMs) have catalyzed significant advancements in Natural Language Processing (NLP), yet they encounter challenges such as hallucination and the need for domain-specific knowledge. To mitigate these, recent methodologies have integrated information retrieved from external resources with LLMs, substantially enhancing their performance across NLP tasks. This survey paper addresses the absence of a comprehensive overview on Retrieval-Augmented Language Models (RALMs), both Retrieval-Augmented Generation (RAG) and Retrieval-Augmented Understanding (RAU), providing an in-depth examination of their paradigm, evolution, taxonomy, and applications. The paper discusses the essential components of RALMs, including Retrievers, Language Models, and Augmentations, and how their interactions lead to diverse model structures and applications. RALMs demonstrate utility in a spectrum of tasks, from translation and dialogue systems to knowledge-intensive applications. The survey includes several evaluation methods of RALMs, emphasizing the importance of robustness, accuracy, and relevance in their assessment. It also acknowledges the limitations of RALMs, particularly in retrieval quality and computational efficiency, offering directions for future research. In conclusion, this survey aims to offer a structured insight into RALMs, their potential, and the avenues for their future development in NLP. The paper is supplemented with a Github Repository containing the surveyed works and resources for further study: https://github.com/2471023025/RALM_Survey.
Large language model pre-training has become increasingly expensive, with most practitioners relying on scaling laws to allocate compute budgets for model size and training tokens, commonly referred to as Compute-Optimal or Chinchilla Optimal. In this paper, we hypothesize a new scaling law that suggests model performance depends mostly on the amount of compute spent for transformer-based models, independent of the specific allocation to model size and dataset size. Using this unified scaling law, we predict that (a) for inference efficiency, training should prioritize smaller model sizes and larger training datasets, and (b) assuming the exhaustion of available web datasets, scaling the model size might be the only way to further improve model performance.
Retrieval-Augmented Generation (RAG) systems are widely used across various industries for querying closed-domain and in-house knowledge bases. However, evaluating these systems presents significant challenges due to the private nature of closed-domain data and a scarcity of queries with verifiable ground truths. Moreover, there is a lack of analytical methods to diagnose problematic modules and identify types of failure, such as those caused by knowledge deficits or issues with robustness. To address these challenges, we introduce GRAMMAR (GRounded And Modular Methodology for Assessment of RAG), an evaluation framework comprising a grounded data generation process and an evaluation protocol that effectively pinpoints defective modules. Our validation experiments reveal that GRAMMAR provides a reliable approach for identifying vulnerable modules and supports hypothesis testing for textual form vulnerabilities. An open-source tool accompanying this framework is available in our GitHub repository (see https://github.com/xinzhel/grammar), allowing for easy reproduction of our results and enabling reliable and modular evaluation in closed-domain settings.
In this paper, we establish a benchmark for table visual question answering, referred to as the TableVQA-Bench, derived from pre-existing table question-answering (QA) and table structure recognition datasets. It is important to note that existing datasets have not incorporated images or QA pairs, which are two crucial components of TableVQA. As such, the primary objective of this paper is to obtain these necessary components. Specifically, images are sourced either through the application of a \textit{stylesheet} or by employing the proposed table rendering system. QA pairs are generated by exploiting the large language model (LLM) where the input is a text-formatted table. Ultimately, the completed TableVQA-Bench comprises 1,500 QA pairs. We comprehensively compare the performance of various multi-modal large language models (MLLMs) on TableVQA-Bench. GPT-4V achieves the highest accuracy among commercial and open-sourced MLLMs from our experiments. Moreover, we discover that the number of vision queries plays a significant role in TableVQA performance. To further analyze the capabilities of MLLMs in comparison to their LLM backbones, we investigate by presenting image-formatted tables to MLLMs and text-formatted tables to LLMs, respectively. Our findings suggest that processing visual inputs is more challenging than text inputs, as evidenced by the lower performance of MLLMs, despite generally requiring higher computational costs than LLMs. The proposed TableVQA-Bench and evaluation codes are available at \href{https://github.com/naver-ai/tablevqabench}{https://github.com/naver-ai/tablevqabench}.
Named Entity Recognition (NER) is an essential steppingstone in the field of natural language processing. Although promising performance has been achieved by various distantly supervised models, we argue that distant supervision inevitably introduces incomplete and noisy annotations, which may mislead the model training process. To address this issue, we propose a robust NER model named BOND-MoE based on Mixture of Experts (MoE). Instead of relying on a single model for NER prediction, multiple models are trained and ensembled under the Expectation-Maximization (EM) framework, so that noisy supervision can be dramatically alleviated. In addition, we introduce a fair assignment module to balance the document-model assignment process. Extensive experiments on real-world datasets show that the proposed method achieves state-of-the-art performance compared with other distantly supervised NER.
Large Language Models (LLMs) have significantly advanced natural language processing (NLP) tasks but also pose ethical and societal risks due to their propensity to generate harmful content. Existing methods have limitations, including the need for training specific control models and proactive intervention during text generation, that lead to quality degradation and increased computational overhead. To mitigate those limitations, we propose LLMSafeGuard, a lightweight real-time framework that integrates an external validator into decoding, rejecting unsafe outputs while allowing valid ones. We introduce a similarity-based validation approach, simplifying constraint introduction and eliminating the need for control model training. Additionally, LLMSafeGuard employs a context-wise timing selection strategy, intervening LLMs only when necessary. We evaluate LLMSafeGuard on detoxification and copyright safeguarding, demonstrating its superiority over SOTA baselines. In detoxification, LLMSafeGuard reduces toxic output by at least 38.6\% while preserving linguistic quality. Additionally, its context-wise timing selection cuts inference time by at least 24.2\% without compromising effectiveness.
Cross-lingual transfer has become an effective way of transferring knowledge between languages. In this paper, we explore an often overlooked aspect in this domain: the influence of the source language of a language model on language transfer performance. We consider a case where the target language and its script are not part of the pre-trained model. We conduct a series of experiments on monolingual and multilingual models that are pre-trained on different tokenization methods to determine factors that affect cross-lingual transfer to a new language with a unique script. Our findings reveal the importance of the tokenizer as a stronger factor than the shared script, language similarity, and model size.
With the proliferation of large language models (LLMs), the comprehensive alignment of such models across multiple tasks has emerged as a critical area of research. Existing alignment methodologies primarily address single task, such as multi-turn dialogue, coding, mathematical problem-solving, and tool usage. However, AI-driven products that leverage language models usually necessitate a fusion of these abilities to function effectively in real-world scenarios. Moreover, the considerable computational resources required for proper alignment of LLMs underscore the need for a more robust, efficient, and encompassing approach to multi-task alignment, ensuring improved generative performance. In response to these challenges, we introduce a novel technique termed Mixture-of-Instructions (MoI), which employs a strategy of instruction concatenation combined with diverse system prompts to boost the alignment efficiency of language models. We have also compiled a diverse set of seven benchmark datasets to rigorously evaluate the alignment efficacy of the MoI-enhanced language model. Our methodology was applied to the open-source Qwen-7B-chat model, culminating in the development of Qwen-SFT-MoI. This enhanced model demonstrates significant advancements in generative capabilities across coding, mathematics, and tool use tasks.
The task of Spell Correction(SC) in low-resource languages presents a significant challenge due to the availability of only a limited corpus of data and no annotated spelling correction datasets. To tackle these challenges a small-scale word-based transformer LM is trained to provide the SC model with contextual understanding. Further, the probabilistic error rules are extracted from the corpus in an unsupervised way to model the tendency of error happening(error model). Then the combination of LM and error model is used to develop the SC model through the well-known noisy channel framework. The effectiveness of this approach is demonstrated through experiments on the Nepali language where there is access to just an unprocessed corpus of textual data.
Multi-choice Machine Reading Comprehension (MMRC) aims to select the correct answer from a set of options based on a given passage and question. The existing methods employ the pre-trained language model as the encoder, share and transfer knowledge through fine-tuning.These methods mainly focus on the design of exquisite mechanisms to effectively capture the relationships among the triplet of passage, question and answers. It is non-trivial but ignored to transfer knowledge from other MRC tasks such as SQuAD due to task specific of MMRC.In this paper, we reconstruct multi-choice to single-choice by training a binary classification to distinguish whether a certain answer is correct. Then select the option with the highest confidence score as the final answer. Our proposed method gets rid of the multi-choice framework and can leverage resources of other tasks. We construct our model based on the ALBERT-xxlarge model and evaluate it on the RACE and DREAM datasets. Experimental results show that our model performs better than multi-choice methods. In addition, by transferring knowledge from other kinds of MRC tasks, our model achieves state-of-the-art results in both single and ensemble settings.
The representation of events in text plays a significant role in various NLP tasks. Recent research demonstrates that contrastive learning has the ability to improve event comprehension capabilities of Pre-trained Language Models (PLMs) and enhance the performance of event representation learning. However, the efficacy of event representation learning based on contrastive learning and PLMs is limited by the short length of event texts. The length of event texts differs significantly from the text length used in the pre-training of PLMs. As a result, there is inconsistency in the distribution of text length between pre-training and event representation learning, which may undermine the learning process of event representation based on PLMs. In this study, we present PromptCL, a novel framework for event representation learning that effectively elicits the capabilities of PLMs to comprehensively capture the semantics of short event texts. PromptCL utilizes a Prompt template borrowed from prompt learning to expand the input text during Contrastive Learning. This helps in enhancing the event representation learning by providing a structured outline of the event components. Moreover, we propose Subject-Predicate-Object (SPO) word order and Event-oriented Masked Language Modeling (EventMLM) to train PLMs to understand the relationships between event components. Our experimental results demonstrate that PromptCL outperforms state-of-the-art baselines on event related tasks. Additionally, we conduct a thorough analysis and demonstrate that using a prompt results in improved generalization capabilities for event representations. Our code will be available at https://github.com/YuboFeng2023/PromptCL.
Prevalent solution for BioNER involves using representation learning techniques coupled with sequence labeling. However, such methods are inherently task-specific, demonstrate poor generalizability, and often require dedicated model for each dataset. To leverage the versatile capabilities of recently remarkable large language models (LLMs), several endeavors have explored generative approaches to entity extraction. Yet, these approaches often fall short of the effectiveness of previouly sequence labeling approaches. In this paper, we utilize the open-sourced LLM LLaMA2 as the backbone model, and design specific instructions to distinguish between different types of entities and datasets. By combining the LLM's understanding of instructions with sequence labeling techniques, we use mix of datasets to train a model capable of extracting various types of entities. Given that the backbone LLMs lacks specialized medical knowledge, we also integrate external entity knowledge bases and employ instruction tuning to compel the model to densely recognize carefully curated entities. Our model VANER, trained with a small partition of parameters, significantly outperforms previous LLMs-based models and, for the first time, as a model based on LLM, surpasses the majority of conventional state-of-the-art BioNER systems, achieving the highest F1 scores across three datasets.
Byte Pair Encoding (BPE) serves as a foundation method for text tokenization in the Natural Language Processing (NLP) field. Despite its wide adoption, the original BPE algorithm harbors an inherent flaw: it inadvertently introduces a frequency imbalance for tokens in the text corpus. Since BPE iteratively merges the most frequent token pair in the text corpus to generate a new token and keeps all generated tokens in the vocabulary, it unavoidably holds tokens that primarily act as components of a longer token and appear infrequently on their own. We term such tokens as Scaffold Tokens. Due to their infrequent occurrences in the text corpus, Scaffold Tokens pose a learning imbalance issue. To address that issue, we propose Scaffold-BPE, which incorporates a dynamic scaffold token removal mechanism by parameter-free, computation-light, and easy-to-implement modifications to the original BPE method. This novel approach ensures the exclusion of low-frequency Scaffold Tokens from the token representations for given texts, thereby mitigating the issue of frequency imbalance and facilitating model training. On extensive experiments across language modeling and even machine translation, Scaffold-BPE consistently outperforms the original BPE, well demonstrating its effectiveness.