We present Superlim, a multi-task NLP benchmark and analysis platform for evaluating Swedish language models, a counterpart to the English-language (Super)GLUE suite. We describe the dataset, the tasks, the leaderboard and report the baseline results yielded by a reference implementation. The tested models do not approach ceiling performance on any of the tasks, which suggests that Superlim is truly difficult, a desirable quality for a benchmark. We address methodological challenges, such as mitigating the Anglocentric bias when creating datasets for a less-resourced language; choosing the most appropriate measures; documenting the datasets and making the leaderboard convenient and transparent. We also highlight other potential usages of the dataset, such as, for instance, the evaluation of cross-lingual transfer learning.
论文检索
输入标题、作者或关键词，从 2,106 篇学术成果中精准定位
Generating concise summaries of news events is a challenging natural language processing task. While journalists often curate timelines to highlight key sub-events, newcomers to a news event face challenges in catching up on its historical context. In this paper, we address this need by introducing the task of background news summarization, which complements each timeline update with a background summary of relevant preceding events. We construct a dataset by merging existing timeline datasets and asking human annotators to write a background summary for each timestep of each news event. We establish strong baseline performance using state-of-the-art summarization systems and propose a query-focused variant to generate background summaries. To evaluate background summary quality, we present a question-answering-based evaluation metric, Background Utility Score (BUS), which measures the percentage of questions about a current event timestep that a background summary answers. Our experiments show the effectiveness of instruction fine-tuned systems such as Flan-T5, in addition to strong zero-shot performance using GPT-3.5.
In multimodal understanding tasks, visual and linguistic ambiguities can arise. Visual ambiguity can occur when visual objects require a model to ground a referring expression in a video without strong supervision, while linguistic ambiguity can occur from changes in entities in action flows. As an example from the cooking domain, “oil” mixed with “salt” and “pepper” could later be referred to as a “mixture”. Without a clear visual-linguistic alignment, we cannot know which among several objects shown is referred to by the language expression “mixture”, and without resolved antecedents, we cannot pinpoint what the mixture is. We define this chicken-and-egg problem as Visual-linguistic Ambiguity. In this paper, we present Find2Find, a joint anaphora resolution and object localization dataset targeting the problem of visual-linguistic ambiguity, consisting of 500 anaphora-annotated recipes with corresponding videos. We present experimental results of a novel end-to-end joint multitask learning framework for Find2Find that fuses visual and textual information and shows improvements both for anaphora resolution and object localization with one joint model in multitask learning, as compared to a strong single-task baseline.
Nearest neighbor machine translation (kNN-MT), which interpolates target token probabilities with estimates derived from additional examples, has achieved significant improvements and attracted extensive interest in recent years. However, existing research does not explicitly consider the source context when retrieving similar examples, potentially leading to suboptimal performance. To address this, we comprehensively revisit the role of source context and propose a simple and effective method for improving neural machine translation via source context enhancement, demonstrating its crucial role in both retrieving superior examples and determining more suitable interpolation coefficients. Furthermore, we reveal that the probability estimation can be further optimized by incorporating a source-aware distance calibration module. Comprehensive experiments show that our proposed approach can be seamlessly integrated with representative kNN-MT baselines, resulting in substantial improvements over these strong baselines across a number of settings and domains. Remarkably, these improvements can reach up to 1.6 BLEU points.
Studying language models (LMs) in terms of well-understood formalisms allows us to precisely characterize their abilities and limitations. Previous work has investigated the expressive power of recurrent neural network (RNN) LMs in terms of their capacity to recognize unweighted formal languages. However, LMs do not describe unweighted formal languages—rather, they define probability distributions over strings. In this work, we study what classes of such probability distributions RNN LMs can represent, which allows us to make more direct statements about their capabilities. We show that simple RNNs are equivalent to a subclass of probabilistic finite-state automata, and can thus model a strict subset of probability distributions expressible by finite-state models. Furthermore, we study the space complexity of representing finite-state LMs with RNNs. We show that, to represent an arbitrary deterministic finite-state LM with N states over an alphabet \Sigma, an RNN requires \Omega\left(N |\Sigma|\right) neurons. These results present a first step towards characterizing the classes of distributions RNN LMs can represent and thus help us understand their capabilities and limitations.
Chain-of-Thought (CoT) prompting has successfully enhanced the reasoning capabilities of Large Language Models (LLMs) with at least 100 billion parameters. However, it is ineffective, or even detrimental, to the performance on reasoning tasks in Smaller Language Models (SLMs) with less than 10 billion parameters. In this paper, we propose Dialogue-guided Chain-of-Thought (DialCoT) to improve the reasoning capabilities of SLMs, with the aim of generating intermediate reasoning steps in a dialogue format to guide the model to the final answer. Furthermore, we optimize the model to choose the optimal reasoning path through the Proximal Policy Optimization (PPO) algorithm, further enhancing its reasoning capabilities. Compared to previous methods, our advantages lie in: 1) We transform the process of solving complex reasoning problems into decomposing problems and solving a series of simpler sub-questions, significantly reducing task difficulty and making it more suitable for SLMs. 2) We optimize the model to choose the optimal reasoning path through the PPO algorithm. Comprehensive experiments on four arithmetic reasoning datasets show that our method can achieve significant performance gains over state-of-the-art competitors.
Our work addresses the problem of unsupervised Aspect Category Detection using a small set of seed words. Recent works have focused on learning embedding spaces for seed words and sentences to establish similarities between sentences and aspects. However, aspect representations are limited by the quality of initial seed words, and model performances are compromised by noise. To mitigate this limitation, we propose a simple framework that automatically enhances the quality of initial seed words and selects high-quality sentences for training instead of using the entire dataset. Our main concepts are to add a number of seed words to the initial set and to treat the task of noise resolution as a task of augmenting data for a low-resource task. In addition, we jointly train Aspect Category Detection with Aspect Term Extraction and Aspect Term Polarity to further enhance performance. This approach facilitates shared representation learning, allowing Aspect Category Detection to benefit from the additional guidance offered by other tasks. Extensive experiments demonstrate that our framework surpasses strong baselines on standard datasets.
New Intent Discovery (NID) aims to recognize both new and known intents from unlabeled data with the aid of limited labeled data containing only known intents. Without considering structure relationships between samples, previous methods generate noisy supervisory signals which cannot strike a balance between quantity and quality, hindering the formation of new intent clusters and effective transfer of the pre-training knowledge. To mitigate this limitation, we propose a novel \textit{Diffusion Weighted Graph Framework} (DWGF) to capture both semantic similarities and structure relationships inherent in data, enabling more sufficient and reliable supervisory signals. Specifically, for each sample, we diffuse neighborhood relationships along semantic paths guided by the nearest neighbors for multiple hops to characterize its local structure discriminately. Then, we sample its positive keys and weigh them based on semantic similarities and local structures for contrastive learning. During inference, we further propose \textit{Graph Smoothing Filter} (GSF) to explicitly utilize the structure relationships to filter high-frequency noise embodied in semantically ambiguous samples on the cluster boundary. Extensive experiments show that our method outperforms state-of-the-art models on all evaluation metrics across multiple benchmark datasets. Code and data will be made public.
Previous approaches for automatic lay summarisation are exclusively reliant on the source article that, given it is written for a technical audience (e.g., researchers), is unlikely to explicitly define all technical concepts or state all of the background information that is relevant for a lay audience. We address this issue by augmenting eLife, an existing biomedical lay summarisation dataset, with article-specific knowledge graphs, each containing detailed information on relevant biomedical concepts. Using both automatic and human evaluations, we systematically investigate the effectiveness of three different approaches for incorporating knowledge graphs within lay summarisation models, with each method targeting a distinct area of the encoder-decoder model architecture. Our results confirm that integrating graph-based domain knowledge can significantly benefit lay summarisation by substantially increasing the readability of generated text and improving the explanation of technical concepts.
Multi-level Adaptive Contrastive Learning for Knowledge Internalization in Dialogue Generation
PDF ↗Knowledge-grounded dialogue generation aims to mitigate the issue of text degeneration by incorporating external knowledge to supplement the context. However, the model often fails to internalize this information into responses in a human-like manner. Instead, it simply inserts segments of the provided knowledge into generic responses. As a result, the generated responses tend to be tedious, incoherent, and in lack of interactivity which means the degeneration problem is still unsolved. In this work, we first find that such copying-style degeneration is primarily due to the weak likelihood objective, which allows the model to “cheat” the objective by merely duplicating knowledge segments in a superficial pattern matching based on overlap. To overcome this challenge, we then propose a Multi-level Adaptive Contrastive Learning (MACL) framework that dynamically samples negative examples and subsequently penalizes degeneration behaviors at both the token-level and sequence-level. Extensive experiments on the WoW dataset demonstrate the effectiveness of our approach across various pre-trained models and decoding strategies.
With the increasing integration of AI into everyday life, it’s becoming crucial to design AI systems to serve users from diverse backgrounds by making them culturally aware. In this paper, we present GD-COMET, a geo-diverse version of the COMET commonsense inference model. GD-COMET goes beyond Western commonsense knowledge and is capable of generating inferences pertaining to a broad range of cultures. We demonstrate the effectiveness of GD-COMET through a comprehensive human evaluation across 5 diverse cultures, as well as extrinsic evaluation on a geo-diverse task. The evaluation shows that GD-COMET captures and generates culturally nuanced commonsense knowledge, demonstrating its potential to benefit NLP applications across the board and contribute to making NLP more inclusive.
Despite the remarkable ability of large language models (LMs) to comprehend and generate language, they have a tendency to hallucinate and create factually inaccurate output. Augmenting LMs by retrieving information from external knowledge resources is one promising solution. Most existing retrieval augmented LMs employ a retrieve-and-generate setup that only retrieves information once based on the input. This is limiting, however, in more general scenarios involving generation of long texts, where continually gathering information throughout generation is essential. In this work, we provide a generalized view of active retrieval augmented generation, methods that actively decide when and what to retrieve across the course of the generation. We propose Forward-Looking Active REtrieval augmented generation (FLARE), a generic method which iteratively uses a prediction of the upcoming sentence to anticipate future content, which is then utilized as a query to retrieve relevant documents to regenerate the sentence if it contains low-confidence tokens. We test FLARE along with baselines comprehensively over 4 long-form knowledge-intensive generation tasks/datasets. FLARE achieves superior or competitive performance on all tasks, demonstrating the effectiveness of our method.
Large Language Models (LLMs) have shown impressive performance as general purpose agents, but their abilities remain highly dependent on prompts which are hand written with onerous trial-and-error effort. We propose a simple and nonparametric solution to this problem, Prompt Optimization with Textual Gradients (ProTeGi), which is inspired by numerical gradient descent to automatically improve prompts, assuming access to training data and an LLM API. The algorithm uses minibatches of data to form natural language “gradients” that criticize the current prompt, much like how numerical gradients point in the direction of error ascent. The natural language gradients are then “propagated” into the prompt by editing the prompt in the opposite semantic direction of the gradient. These gradient descent steps are guided by a beam search and bandit selection procedure which significantly improves algorithmic efficiency. Preliminary results across three benchmark NLP tasks and the novel problem of LLM jailbreak detection suggest that Automatic Prompt Optimization can outperform prior prompt editing techniques and improve an initial prompt’s performance by up to 31%, by using data to rewrite vague task descriptions into more precise annotation instructions.
Large language models (LLMs) exhibited powerful capability in various natural language processing tasks. This work focuses on exploring LLM performance on zero-shot information extraction, with a focus on the ChatGPT and named entity recognition (NER) task. Inspired by the remarkable reasoning capability of LLM on symbolic and arithmetic reasoning, we adapt the prevalent reasoning methods to NER and propose reasoning strategies tailored for NER. First, we explore a decomposed question-answering paradigm by breaking down the NER task into simpler subproblems by labels. Second, we propose syntactic augmentation to stimulate the model’s intermediate thinking in two ways: syntactic prompting, which encourages the model to analyze the syntactic structure itself, and tool augmentation, which provides the model with the syntactic information generated by a parsing tool. Besides, we adapt self-consistency to NER by proposing a two-stage majority voting strategy, which first votes for the most consistent mentions, then the most consistent types. The proposed methods achieve remarkable improvements for zero-shot NER across seven benchmarks, including Chinese and English datasets, and on both domain-specific and general-domain scenarios. In addition, we present a comprehensive analysis of the error types with suggestions for optimization directions. We also verify the effectiveness of the proposed methods on the few-shot setting and other LLMs.
M^3Seg: A Maximum-Minimum Mutual Information Paradigm for Unsupervised Topic Segmentation in ASR Transcripts
PDF ↗Topic segmentation aims to detect topic boundaries and split automatic speech recognition transcriptions (e.g., meeting transcripts) into segments that are bounded by thematic meanings. In this work, we propose M^3Seg, a novel Maximum-Minimum Mutual information paradigm for linear topic segmentation without using any parallel data. Specifically, by employing sentence representations provided by pre-trained language models, M^3Seg first learns a region-based segment encoder based on the maximization of mutual information between the global segment representation and the local contextual sentence representation. Secondly, an edge-based boundary detection module aims to segment the whole by topics based on minimizing the mutual information between different segments. Experiment results on two public datasets demonstrate the effectiveness of M^3Seg, which outperform the state-of-the-art methods by a significant (18%–37% improvement) margin.
Don’t Trust ChatGPT when your Question is not in English: A Study of Multilingual Abilities and Types of LLMs
PDF ↗Large language models (LLMs) have demonstrated exceptional natural language understanding abilities, and have excelled in a variety of natural language processing (NLP) tasks. Despite the fact that most LLMs are trained predominantly on English, multiple studies have demonstrated their capabilities in a variety of languages. However, fundamental questions persist regarding how LLMs acquire their multilingual abilities and how performance varies across different languages. These inquiries are crucial for the study of LLMs since users and researchers often come from diverse language backgrounds, potentially influencing how they use LLMs and interpret their output. In this work, we propose a systematic way of qualitatively and quantitatively evaluating the multilingual capabilities of LLMs. We investigate the phenomenon of cross-language generalization in LLMs, wherein limited multilingual training data leads to advanced multilingual capabilities. To accomplish this, we employ a novel prompt back-translation method. The results demonstrate that LLMs, such as GPT, can effectively transfer learned knowledge across different languages, yielding relatively consistent results in translation-equivariant tasks, in which the correct output does not depend on the language of the input. However, LLMs struggle to provide accurate results in translation-variant tasks, which lack this property, requiring careful user judgment to evaluate the answers.
Dialogue state error correction has recently been proposed to correct wrong slot values in predicted dialogue states, thereby mitigating the error propagation problem for dialogue state tracking (DST). These approaches, though effective, are heavily intertwined with specific DST models, limiting their applicability to other DST models. To solve this problem, we propose Scalable Dialogue State Correction (Scalable-DSC), which can correct wrong slot values in the dialogue state predicted by any DST model. Specifically, we propose a Structural Template Prompt (STP) that converts predicted dialogue state from any DST models into a standardized natural language sequence as a part of the historical context, associates them with dialogue history information, and generates a corrected dialogue state sequence based on predefined template options. We further enhance Scalable-DSC by introducing two training strategies. The first employs a predictive state simulator to simulate the predicted dialogue states as the training data to enhance the generalization ability of the model. The second involves using the dialogue state predicted by DST as the training data, aiming at mitigating the inconsistent error type distribution between the training and inference. Experiments confirm that our model achieves state-of-the-art results on MultiWOZ 2.0-2.4.
The recent LLMs like GPT-4 and PaLM-2 have made tremendous progress in solving fundamental math problems like GSM8K by achieving over 90% accuracy. However, their capabilities to solve more challenging math problems which require domain-specific knowledge (i.e. theorem) have yet to be investigated. In this paper, we introduce TheoremQA, the first theorem-driven question-answering dataset designed to evaluate AI models’ capabilities to apply theorems to solve challenging science problems. TheoremQA is curated by domain experts containing 800 high-quality questions covering 350 theorems from Math, Physics, EE&CS, and Finance. We evaluate a wide spectrum of 16 large language and code models with different prompting strategies like Chain-of-Thoughts and Program-of-Thoughts. We found that GPT-4’s capabilities to solve these problems are unparalleled, achieving an accuracy of 51% with Program-of-Thoughts Prompting. All the existing open-sourced models are below 15%, barely surpassing the random-guess baseline. Given the diversity and broad coverage of TheoremQA, we believe it can be used as a better benchmark to evaluate LLMs’ capabilities to solve challenging science problems.
Federated Learning of Large Language Models with Parameter-Efficient Prompt Tuning and Adaptive Optimization
PDF ↗Federated learning (FL) is a promising paradigm to enable collaborative model training with decentralized data. However, the training process of Large Language Models (LLMs) generally incurs the update of significant parameters, which limits the applicability of FL techniques to tackle the LLMs in real scenarios. Prompt tuning can significantly reduce the number of parameters to update, but it either incurs performance degradation or low training efficiency. The straightforward utilization of prompt tuning in the FL often raises non-trivial communication costs and dramatically degrades performance. In addition, the decentralized data is generally non-Independent and Identically Distributed (non-IID), which brings client drift problems and thus poor performance. This paper proposes a Parameter-efficient prompt Tuning approach with Adaptive Optimization, i.e., FedPepTAO, to enable efficient and effective FL of LLMs. First, an efficient partial prompt tuning approach is proposed to improve performance and efficiency simultaneously. Second, a novel adaptive optimization method is developed to address the client drift problems on both the device and server sides to enhance performance further. Extensive experiments based on 10 datasets demonstrate the superb performance (up to 60.8% in terms of accuracy) and efficiency (up to 97.59% in terms of training time) of FedPepTAO compared with 9 baseline approaches. Our code is available at https://github.com/llm-eff/FedPepTAO.
Large Language Models (LLMs) are trained on corpora disproportionally weighted in favor of Standard American English. As a result, speakers of other dialects experience significantly more failures when interacting with these technologies. In practice, these speakers often accommodate their speech to be better understood. Our work shares the belief that language technologies should be designed to accommodate the diversity in English dialects and not the other way around. However, prior work on dialect struggle with generalizing to evolving and emerging dialects in a scalable manner. To fill this gap, our method, HyperLoRA, leverages expert linguistic knowledge to enable resource-efficient adaptation via hypernetworks. By disentangling dialect-specific and cross-dialectal information, HyperLoRA improves generalization to unseen dialects in a task-agnostic fashion. Not only is HyperLoRA more scalable in the number of parameters, but it also achieves the best or most competitive performance across 5 dialects in a zero-shot setting. In this way, our approach facilitates access to language technology for billions of English dialect speakers who are traditionally underrepresented.