Oasis: Data Curation and Assessment System for Pretraining of Large Language Models

Data is one of the most critical elements in building a large language model. However, existing systems either fail to customize a corpus curation pipeline or neglect to leverage comprehensive corpus assessment for iterative optimization of the curation. To this end, we present a pretraining corpus curation and assessment platform called Oasis -- a one-stop system for data quality improvement and quantification with user-friendly interactive interfaces. Specifically, the interactive modular rule filter module can devise customized rules according to explicit feedback. The debiased neural filter module builds the quality classification dataset in a negative-centric manner to remove the undesired bias. The adaptive document deduplication module could execute large-scale deduplication with limited memory resources. These three parts constitute the customized data curation module. And in the holistic data assessment module, a corpus can be assessed in local and global views, with three evaluation means including human, GPT-4, and heuristic metrics. We exhibit a complete process to use Oasis for the curation and assessment of pretraining data. In addition, an 800GB bilingual corpus curated by Oasis is publicly released.

Assessing Knowledge Editing in Language Models via Relation Perspective

Knowledge Editing (KE) for modifying factual knowledge in Large Language Models (LLMs) has been receiving increasing attention. However, existing knowledge editing methods are entity-centric, and it is unclear whether this approach is suitable for a relation-centric perspective. To address this gap, this paper constructs a new benchmark named RaKE, which focuses on Relation based Knowledge Editing. In this paper, we establish a suite of innovative metrics for evaluation and conduct comprehensive experiments involving various knowledge editing baselines. We notice that existing knowledge editing methods exhibit the potential difficulty in their ability to edit relations. Therefore, we further explore the role of relations in factual triplets within the transformer. Our research results confirm that knowledge related to relations is not only stored in the FFN network but also in the attention layers. This provides experimental support for future relation-based knowledge editing methods.

ExpNote: Black-box Large Language Models are Better Task Solvers with Experience Notebook

Black-box Large Language Models (LLMs) have shown great power in solving various tasks and are considered general problem solvers. However, LLMs still fail in many specific tasks although understand the task instruction. In this paper, we focus on the problem of boosting the ability of black-box LLMs to solve downstream tasks. We propose ExpNote, an automated framework to help LLMs better adapt to unfamiliar tasks through reflecting and noting experiences from training data and retrieving them from external memory during testing. We evaluate ExpNote on multiple tasks and the experimental results demonstrate that the proposed method significantly improves the performance of black-box LLMs. The data and code are available at https://github.com/forangel2014/ExpNote

S3Eval: A Synthetic, Scalable, Systematic Evaluation Suite for Large Language Models

The rapid development of Large Language Models (LLMs) has led to great strides in model capabilities like reasoning and long-context understanding. However, as LLMs are able to process longer contexts, it becomes more challenging to evaluate whether they have acquired certain capabilities, since the length of text (e.g., 100K tokens) they can process far exceeds what humans can reliably assess in a reasonable duration. In this paper, we propose using complex synthetic tasks as a proxy evaluation method, and present S3Eval, a Synthetic, Scalable, Systematic evaluation suite for LLMs evaluation. As a synthetic benchmark, S3Eval enables the creation of any number of evaluation examples that are theoretically invisible to LLMs, mitigating the test set contamination issue. The synthetic nature of S3Eval provides users full control over the dataset, allowing them to systematically probe LLM capabilities by scaling text length and varying task difficulty across diverse scenarios. The strong correlation between S3Eval performance and scores of real-world benchmarks like Big-Bench Hard (BBH) demonstrates the soundness of using S3Eval for evaluation of LLMs. The in-depth analysis also uncover additional insights, including performance drop when the answer is sparsely distributed or located in the middle context, as well as some counter-intuitive trends of model performance.

TableQAKit: A Comprehensive and Practical Toolkit for Table-based Question Answering

Table-based question answering (TableQA) is an important task in natural language processing, which requires comprehending tables and employing various reasoning ways to answer the questions. This paper introduces TableQAKit, the first comprehensive toolkit designed specifically for TableQA. The toolkit designs a unified platform that includes plentiful TableQA datasets and integrates popular methods of this task as well as large language models (LLMs). Users can add their datasets and methods according to the friendly interface. Also, pleasantly surprised using the modules in this toolkit achieves new SOTA on some datasets. Finally, \tableqakit{} also provides an LLM-based TableQA Benchmark for evaluating the role of LLMs in TableQA. TableQAKit is open-source with an interactive interface that includes visual operations, and comprehensive data for ease of use.

Query2Triple: Unified Query Encoding for Answering Diverse Complex Queries over Knowledge Graphs

Complex Query Answering (CQA) is a challenge task of Knowledge Graph (KG). Due to the incompleteness of KGs, query embedding (QE) methods have been proposed to encode queries and entities into the same embedding space, and treat logical operators as neural set operators to obtain answers. However, these methods train KG embeddings and neural set operators concurrently on both simple (one-hop) and complex (multi-hop and logical) queries, which causes performance degradation on simple queries and low training efficiency. In this paper, we propose Query to Triple (Q2T), a novel approach that decouples the training for simple and complex queries. Q2T divides the training into two stages: (1) Pre-training a neural link predictor on simple queries to predict tail entities based on the head entity and relation. (2) Training a query encoder on complex queries to encode diverse complex queries into a unified triple form that can be efficiently solved by the pretrained neural link predictor. Our proposed Q2T is not only efficient to train, but also modular, thus easily adaptable to various neural link predictors that have been studied well. Extensive experiments demonstrate that, even without explicit modeling for neural set operators, Q2T still achieves state-of-the-art performance on diverse complex queries over three public benchmarks.

Generative Calibration for In-context Learning

As one of the most exciting features of large language models (LLMs), in-context learning is a mixed blessing. While it allows users to fast-prototype a task solver with only a few training examples, the performance is generally sensitive to various configurations of the prompt such as the choice or order of the training examples. In this paper, we for the first time theoretically and empirically identify that such a paradox is mainly due to the label shift of the in-context model to the data distribution, in which LLMs shift the label marginal $p(y)$ while having a good label conditional $p(x|y)$. With this understanding, we can simply calibrate the in-context predictive distribution by adjusting the label marginal, which is estimated via Monte-Carlo sampling over the in-context model, i.e., generation of LLMs. We call our approach as generative calibration. We conduct exhaustive experiments with 12 text classification tasks and 12 LLMs scaling from 774M to 33B, generally find that the proposed method greatly and consistently outperforms the ICL as well as state-of-the-art calibration methods, by up to 27% absolute in macro-F1. Meanwhile, the proposed method is also stable under different prompt configurations.

MenatQA: A New Dataset for Testing the Temporal Comprehension and Reasoning Abilities of Large Language Models

Large language models (LLMs) have shown nearly saturated performance on many natural language processing (NLP) tasks. As a result, it is natural for people to believe that LLMs have also mastered abilities such as time understanding and reasoning. However, research on the temporal sensitivity of LLMs has been insufficiently emphasized. To fill this gap, this paper constructs Multiple Sensitive Factors Time QA (MenatQA), which encompasses three temporal factors (scope factor, order factor, counterfactual factor) with total 2,853 samples for evaluating the time comprehension and reasoning abilities of LLMs. This paper tests current mainstream LLMs with different parameter sizes, ranging from billions to hundreds of billions. The results show most LLMs fall behind smaller temporal reasoning models with different degree on these factors. In specific, LLMs show a significant vulnerability to temporal biases and depend heavily on the temporal information provided in questions. Furthermore, this paper undertakes a preliminary investigation into potential improvement strategies by devising specific prompts and leveraging external tools. These approaches serve as valuable baselines or references for future research endeavors.

HRoT: Hybrid prompt strategy and Retrieval of Thought for Table-Text Hybrid Question Answering

Answering numerical questions over hybrid contents from the given tables and text(TextTableQA) is a challenging task. Recently, Large Language Models (LLMs) have gained significant attention in the NLP community. With the emergence of large language models, In-Context Learning and Chain-of-Thought prompting have become two particularly popular research topics in this field. In this paper, we introduce a new prompting strategy called Hybrid prompt strategy and Retrieval of Thought for TextTableQA. Through In-Context Learning, we prompt the model to develop the ability of retrieval thinking when dealing with hybrid data. Our method achieves superior performance compared to the fully-supervised SOTA on the MultiHiertt dataset in the few-shot setting.

MMHQA-ICL: Multimodal In-context Learning for Hybrid Question Answering over Text, Tables and Images

In the real world, knowledge often exists in a multimodal and heterogeneous form. Addressing the task of question answering with hybrid data types, including text, tables, and images, is a challenging task (MMHQA). Recently, with the rise of large language models (LLM), in-context learning (ICL) has become the most popular way to solve QA problems. We propose MMHQA-ICL framework for addressing this problems, which includes stronger heterogeneous data retriever and an image caption module. Most importantly, we propose a Type-specific In-context Learning Strategy for MMHQA, enabling LLMs to leverage their powerful performance in this task. We are the first to use end-to-end LLM prompting method for this task. Experimental results demonstrate that our framework outperforms all baselines and methods trained on the full dataset, achieving state-of-the-art results under the few-shot setting on the MultimodalQA dataset.