Large language and multimodal models have shown remarkable successes on various benchmarks focused on specific skills such as general-purpose programming, natural language understanding, math word problem-solving, and visual question answering. However, it is unclear how well these models perform on tasks that require a combination of these skills. In this paper, we curate a novel program synthesis benchmark based on the XLogoOnline visual programming environment. The benchmark comprises 85 real-world tasks from the Mini-level of the XLogoOnline environment, each requiring a combination of different skills such as spatial planning, basic programming, and logical reasoning. Our evaluation shows that current state-of-the-art models like GPT-4V and Llama3-70B struggle to solve these tasks, achieving only 20% and 2.35% success rates. Next, we develop a fine-tuning pipeline to boost the performance of models by leveraging a large-scale synthetic training dataset with over 80000 tasks. Moreover, we showcase how emulator-driven feedback can be used to design a curriculum over training data distribution. We showcase that a fine-tuned Llama3-8B drastically outperforms GPT-4V and Llama3-70B models, and provide an in-depth analysis of the models' expertise across different skill dimensions. We will publicly release the benchmark for future research on program synthesis in visual programming.
Inspired by the Bloch Sphere representation, we propose a novel rotary position encoding on a three-dimensional sphere, named 3D Rotary Position Encoding (3D-RPE). 3D-RPE is an advanced version of the widely used 2D Rotary Position Encoding (RoPE), with two major advantages for modeling long contexts: controllable long-term decay and improved position resolution. For controllable long-term decay, 3D-RPE allows for the regulation of long-term decay within the chunk size, ensuring the modeling of relative positional information between tokens at a distant relative position. For enhanced position resolution, 3D-RPE can mitigate the degradation of position resolution caused by position interpolation on RoPE. We have conducted experiments on long-context Natural Language Understanding (NLU) and long-sequence Language Modeling (LM) tasks. From the experimental results, 3D-RPE achieved performance improvements over RoPE, especially in long-context NLU tasks.
Recent language models have shown remarkable performance on natural language understanding (NLU) tasks. However, they are often sub-optimal when faced with ambiguous samples that can be interpreted in multiple ways, over-confidently predicting a single label without consideration for its correctness. To address this issue, we propose a novel self-knowledge distillation method that enables models to learn label distributions more accurately by leveraging knowledge distilled from their lower layers. This approach also includes a learning phase that re-calibrates the unnecessarily strengthened confidence for training samples judged as extremely ambiguous based on the distilled distribution knowledge. We validate our method on diverse NLU benchmark datasets and the experimental results demonstrate its effectiveness in producing better label distributions. Particularly, through the process of re-calibrating the confidence for highly ambiguous samples, the issue of over-confidence when predictions for unseen samples do not match with their ground-truth labels has been significantly alleviated. This has been shown to contribute to generating better distributions than the existing state-of-the-art method. Moreover, our method is more efficient in training the models compared to the existing method, as it does not involve additional training processes to refine label distributions.
Transformers have revolutionized machine learning with their simple yet effective architecture. Pre-training Transformers on massive text datasets from the Internet has led to unmatched generalization for natural language understanding (NLU) tasks. However, such language models remain fragile when tasked with algorithmic forms of reasoning, where computations must be precise and robust. To address this limitation, we propose a novel approach that combines the Transformer's language understanding with the robustness of graph neural network (GNN)-based neural algorithmic reasoners (NARs). Such NARs proved effective as generic solvers for algorithmic tasks, when specified in graph form. To make their embeddings accessible to a Transformer, we propose a hybrid architecture with a two-phase training procedure, allowing the tokens in the language model to cross-attend to the node embeddings from the NAR. We evaluate our resulting TransNAR model on CLRS-Text, the text-based version of the CLRS-30 benchmark, and demonstrate significant gains over Transformer-only models for algorithmic reasoning, both in and out of distribution.
Crosswords are a form of word puzzle that require a solver to demonstrate a high degree of proficiency in natural language understanding, wordplay, reasoning, and world knowledge, along with adherence to character and length constraints. In this paper we tackle the challenge of solving crosswords with Large Language Models (LLMs). We demonstrate that the current generation of state-of-the art (SoTA) language models show significant competence at deciphering cryptic crossword clues, and outperform previously reported SoTA results by a factor of 2-3 in relevant benchmarks. We also develop a search algorithm that builds off this performance to tackle the problem of solving full crossword grids with LLMs for the very first time, achieving an accuracy of 93\% on New York Times crossword puzzles. Contrary to previous work in this area which concluded that LLMs lag human expert performance significantly, our research suggests this gap is a lot narrower.
Generating accurate SQL from users' natural language questions (text-to-SQL) remains a long-standing challenge due to the complexities involved in user question understanding, database schema comprehension, and SQL generation. Traditional text-to-SQL systems, which combine human engineering and deep neural networks, have made significant progress. Subsequently, pre-trained language models (PLMs) have been developed for text-to-SQL tasks, achieving promising results. However, as modern databases and user questions grow more complex, PLMs with a limited parameter size often produce incorrect SQL. This necessitates more sophisticated and tailored optimization methods, which restricts the application of PLM-based systems. Recently, large language models (LLMs) have shown significant capabilities in natural language understanding as model scale increases. Thus, integrating LLM-based solutions can bring unique opportunities, improvements, and solutions to text-to-SQL research. In this survey, we provide a comprehensive review of existing LLM-based text-to-SQL studies. Specifically, we offer a brief overview of the technical challenges and evolutionary process of text-to-SQL. Next, we introduce the datasets and metrics designed to evaluate text-to-SQL systems. Subsequently, we present a systematic analysis of recent advances in LLM-based text-to-SQL. Finally, we make a summarization and discuss the remaining challenges in this field and suggest expectations for future research directions.
Indirect User Requests (IURs), such as "It's cold in here" instead of "Could you please increase the temperature?" are common in human-human task-oriented dialogue and require world knowledge and pragmatic reasoning from the listener. While large language models (LLMs) can handle these requests effectively, smaller models deployed on virtual assistants often struggle due to resource constraints. Moreover, existing task-oriented dialogue benchmarks lack sufficient examples of complex discourse phenomena such as indirectness. To address this, we propose a set of linguistic criteria along with an LLM-based pipeline for generating realistic IURs to test natural language understanding (NLU) and dialogue state tracking (DST) models before deployment in a new domain. We also release IndirectRequests, a dataset of IURs based on the Schema Guided Dialog (SGD) corpus, as a comparative testbed for evaluating the performance of smaller models in handling indirect requests.
Crowdsourcing is a critical technology in social manufacturing, which leverages an extensive and boundless reservoir of human resources to handle a wide array of complex tasks. The successful execution of these complex tasks relies on task decomposition (TD) and allocation, with the former being a prerequisite for the latter. Recently, pre-trained language models (PLMs)-based methods have garnered significant attention. However, they are constrained to handling straightforward common-sense tasks due to their inherent restrictions involving limited and difficult-to-update knowledge as well as the presence of hallucinations. To address these issues, we propose a retrieval-augmented generation-based crowdsourcing framework that reimagines TD as event detection from the perspective of natural language understanding. However, the existing detection methods fail to distinguish differences between event types and always depend on heuristic rules and external semantic analyzing tools. Therefore, we present a Prompt-Based Contrastive learning framework for TD (PBCT), which incorporates a prompt-based trigger detector to overcome dependence. Additionally, trigger-attentive sentinel and masked contrastive learning are introduced to provide varying attention to trigger and contextual features according to different event types. Experiment results demonstrate the competitiveness of our method in both supervised and zero-shot detection. A case study on printed circuit board manufacturing is showcased to validate its adaptability to unknown professional domains.
A Language Model is a term that encompasses various types of models designed to understand and generate human communication. Large Language Models (LLMs) have gained significant attention due to their ability to process text with human-like fluency and coherence, making them valuable for a wide range of data-related tasks fashioned as pipelines. The capabilities of LLMs in natural language understanding and generation, combined with their scalability, versatility, and state-of-the-art performance, enable innovative applications across various AI-related fields, including eXplainable Artificial Intelligence (XAI), Automated Machine Learning (AutoML), and Knowledge Graphs (KG). Furthermore, we believe these models can extract valuable insights and make data-driven decisions at scale, a practice commonly referred to as Big Data Analytics (BDA). In this position paper, we provide some discussions in the direction of unlocking synergies among these technologies, which can lead to more powerful and intelligent AI solutions, driving improvements in data pipelines across a wide range of applications and domains integrating humans, computers, and knowledge.
Negation is a common linguistic phenomenon. Yet language models face challenges with negation in many natural language understanding tasks such as question answering and natural language inference. In this paper, we experiment with seamless strategies that incorporate affirmative interpretations (i.e., paraphrases without negation) to make models more robust against negation. Crucially, our affirmative interpretations are obtained automatically. We show improvements with CondaQA, a large corpus requiring reasoning with negation, and five natural language understanding tasks.
Large Language Models (LLMs) with their strong task-handling capabilities have shown remarkable advancements across a spectrum of fields, moving beyond natural language understanding. However, their proficiency within the chemistry domain remains restricted, especially in solving professional molecule-related tasks. This challenge is attributed to their inherent limitations in comprehending molecules using only common textual representations, i.e., SMILES strings. In this study, we seek to enhance the ability of LLMs to comprehend molecules by equipping them with a multi-modal external module, namely MolX. In particular, instead of directly using a SMILES string to represent a molecule, we utilize specific encoders to extract fine-grained features from both SMILES string and 2D molecular graph representations for feeding into an LLM. Moreover, a handcrafted molecular fingerprint is incorporated to leverage its embedded domain knowledge. Then, to establish an alignment between MolX and the LLM's textual input space, the whole model in which the LLM is frozen, is pre-trained with a versatile strategy including a diverse set of tasks. Experimental evaluations show that our proposed method outperforms baselines across 4 downstream molecule-related tasks ranging from molecule-to-text translation to retrosynthesis, with and without fine-tuning the LLM, while only introducing a small number of trainable parameters 0.53% and 0.82%, respectively.
Large language models have demonstrated remarkable few-shot performance on many natural language understanding tasks. Despite several demonstrations of using large language models in complex, strategic scenarios, there lacks a comprehensive framework for evaluating agents' performance across various types of reasoning found in games. To address this gap, we introduce GameBench, a cross-domain benchmark for evaluating strategic reasoning abilities of LLM agents. We focus on 9 different game environments, where each covers at least one axis of key reasoning skill identified in strategy games, and select games for which strategy explanations are unlikely to form a significant portion of models' pretraining corpuses. Our evaluations use GPT-3 and GPT-4 in their base form along with two scaffolding frameworks designed to enhance strategic reasoning ability: Chain-of-Thought (CoT) prompting and Reasoning Via Planning (RAP). Our results show that none of the tested models match human performance, and at worst GPT-4 performs worse than random action. CoT and RAP both improve scores but not comparable to human levels.
Recent Pre-trained Language Models (PLMs) usually only provide users with the inference APIs, namely the emerging Model-as-a-Service (MaaS) setting. To adapt MaaS PLMs to downstream tasks without accessing their parameters and gradients, some existing methods focus on the output-side adaptation of PLMs, viewing the PLM as an encoder and then optimizing a task-specific decoder for decoding the output hidden states and class scores of the PLM. Despite the effectiveness of these methods, they only use a single prompt to query PLMs for decoding, leading to a heavy reliance on the quality of the adopted prompt. In this paper, we propose a simple yet effective Multi-Prompting Decoder (MPD) framework for MaaS adaptation. The core idea is to query PLMs with multiple different prompts for each sample, thereby obtaining multiple output hidden states and class scores for subsequent decoding. Such multi-prompting decoding paradigm can simultaneously mitigate reliance on the quality of a single prompt, alleviate the issue of data scarcity under the few-shot setting, and provide richer knowledge extracted from PLMs. Specifically, we propose two decoding strategies: multi-prompting decoding with optimal transport for hidden states and calibrated decoding for class scores. Extensive experiments demonstrate that our method achieves new state-of-the-art results on multiple natural language understanding datasets under the few-shot setting.
A timeline provides a total ordering of events and times, and is useful for a number of natural language understanding tasks. However, qualitative temporal graphs that can be derived directly from text -- such as TimeML annotations -- usually explicitly reveal only partial orderings of events and times. In this work, we apply prior work on solving point algebra problems to the task of extracting timelines from TimeML annotated texts, and develop an exact, end-to-end solution which we call TLEX (TimeLine EXtraction). TLEX transforms TimeML annotations into a collection of timelines arranged in a trunk-and-branch structure. Like what has been done in prior work, TLEX checks the consistency of the temporal graph and solves it; however, it adds two novel functionalities. First, it identifies specific relations involved in an inconsistency (which could then be manually corrected) and, second, TLEX performs a novel identification of sections of the timelines that have indeterminate order, information critical for downstream tasks such as aligning events from different timelines. We provide detailed descriptions and analysis of the algorithmic components in TLEX, and conduct experimental evaluations by applying TLEX to 385 TimeML annotated texts from four corpora. We show that 123 of the texts are inconsistent, 181 of them have more than one ``real world'' or main timeline, and there are 2,541 indeterminate sections across all four corpora. A sampling evaluation showed that TLEX is 98--100% accurate with 95% confidence along five dimensions: the ordering of time-points, the number of main timelines, the placement of time-points on main versus subordinate timelines, the connecting point of branch timelines, and the location of the indeterminate sections. We provide a reference implementation of TLEX, the extracted timelines for all texts, and the manual corrections of the inconsistent texts.
Recent advances in large language models (LLMs) demonstrate substantial capabilities in natural language understanding and generation tasks. With the growing number of LLMs, how to harness the collective expertise of multiple LLMs is an exciting open direction. Toward this goal, we propose a new approach that leverages the collective strengths of multiple LLMs through a Mixture-of-Agents (MoA) methodology. In our approach, we construct a layered MoA architecture wherein each layer comprises multiple LLM agents. Each agent takes all the outputs from agents in the previous layer as auxiliary information in generating its response. MoA models achieves state-of-art performance on AlpacaEval 2.0, MT-Bench and FLASK, surpassing GPT-4 Omni. For example, our MoA using only open-source LLMs is the leader of AlpacaEval 2.0 by a substantial gap, achieving a score of 65.1% compared to 57.5% by GPT-4 Omni.
We present ABEX, a novel and effective generative data augmentation methodology for low-resource Natural Language Understanding (NLU) tasks. ABEX is based on ABstract-and-EXpand, a novel paradigm for generating diverse forms of an input document -- we first convert a document into its concise, abstract description and then generate new documents based on expanding the resultant abstraction. To learn the task of expanding abstract descriptions, we first train BART on a large-scale synthetic dataset with abstract-document pairs. Next, to generate abstract descriptions for a document, we propose a simple, controllable, and training-free method based on editing AMR graphs. ABEX brings the best of both worlds: by expanding from abstract representations, it preserves the original semantic properties of the documents, like style and meaning, thereby maintaining alignment with the original label and data distribution. At the same time, the fundamental process of elaborating on abstract descriptions facilitates diverse generations. We demonstrate the effectiveness of ABEX on 4 NLU tasks spanning 12 datasets and 4 low-resource settings. ABEX outperforms all our baselines qualitatively with improvements of 0.04% - 38.8%. Qualitatively, ABEX outperforms all prior methods from literature in terms of context and length diversity.
It is very challenging to curate a dataset for language-specific knowledge and common sense in order to evaluate natural language understanding capabilities of language models. Due to the limitation in the availability of annotators, most current multilingual datasets are created through translation, which cannot evaluate such language-specific aspects. Therefore, we propose Multilingual CommonsenseQA (mCSQA) based on the construction process of CSQA but leveraging language models for a more efficient construction, e.g., by asking LM to generate questions/answers, refine answers and verify QAs followed by reduced human efforts for verification. Constructed dataset is a benchmark for cross-lingual language-transfer capabilities of multilingual LMs, and experimental results showed high language-transfer capabilities for questions that LMs could easily solve, but lower transfer capabilities for questions requiring deep knowledge or commonsense. This highlights the necessity of language-specific datasets for evaluation and training. Finally, our method demonstrated that multilingual LMs could create QA including language-specific knowledge, significantly reducing the dataset creation cost compared to manual creation. The datasets are available at https://huggingface.co/datasets/yusuke1997/mCSQA.
Discourse Representation Structure (DRS) is an innovative semantic representation designed to capture the meaning of texts with arbitrary lengths across languages. The semantic representation parsing is essential for achieving natural language understanding through logical forms. Nevertheless, the performance of DRS parsing models remains constrained when trained exclusively on monolingual data. To tackle this issue, we introduce a cross-lingual training strategy. The proposed method is model-agnostic yet highly effective. It leverages cross-lingual training data and fully exploits the alignments between languages encoded in pre-trained language models. The experiments conducted on the standard benchmarks demonstrate that models trained using the cross-lingual training method exhibit significant improvements in DRS clause and graph parsing in English, German, Italian and Dutch. Comparing our final models to previous works, we achieve state-of-the-art results in the standard benchmarks. Furthermore, the detailed analysis provides deep insights into the performance of the parsers, offering inspiration for future research in DRS parsing. We keep updating new results on benchmarks to the appendix.
Large language models (LLMs) exhibit impressive proficiency in natural language generation, understanding user instructions, and emulating human-like language use, which has led to significant interest in their application to role-playing scenarios. However, the manual collection of role-specific script data and the evaluation of model performance are resource-intensive processes. This paper introduces a prompt-based framework designed to leverage GPT's capabilities for the generation of role-playing dialogue datasets and the evaluation of role-playing performance. To validate the effectiveness of the GPT-based generation and evaluation, we further incorporate the recall-oriented Rouge-L metric, providing an additional quantitative measure of performance.
Recent strides in large language models (LLMs) have yielded remarkable performance, leveraging reinforcement learning from human feedback (RLHF) to significantly enhance generation and alignment capabilities. However, RLHF encounters numerous challenges, including the objective mismatch issue, leading to suboptimal performance in Natural Language Understanding (NLU) tasks. To address this limitation, we propose a novel Reinforcement Learning framework enhanced with Label-sensitive Reward (RLLR) to amplify the performance of LLMs in NLU tasks. By incorporating label-sensitive pairs into reinforcement learning, our method aims to adeptly capture nuanced label-sensitive semantic features during RL, thereby enhancing natural language understanding. Experiments conducted on five diverse foundation models across eight tasks showcase promising results. In comparison to Supervised Fine-tuning models (SFT), RLLR demonstrates an average performance improvement of 1.54%. Compared with RLHF models, the improvement averages at 0.69%. These results reveal the effectiveness of our method for LLMs in NLU tasks. Code and data available at: https://github.com/MagiaSN/ACL2024_RLLR.