Zheng Yuan

Zheng Yuan, Hongyi Yuan, Chuanqi Tan et al. · tsinghua

Reinforcement Learning from Human Feedback (RLHF) facilitates the alignment of large language models with human preferences, significantly enhancing the quality of interactions between humans and models. InstructGPT implements RLHF through several stages, including Supervised Fine-Tuning (SFT), reward model training, and Proximal Policy Optimization (PPO). However, PPO is sensitive to hyperparameters and requires multiple models in its standard implementation, making it hard to train and scale up to larger parameter counts. In contrast, we propose a novel learning paradigm called RRHF, which scores sampled responses from different sources via a logarithm of conditional probabilities and learns to align these probabilities with human preferences through ranking loss. RRHF can leverage sampled responses from various sources including the model responses from itself, other large language model responses, and human expert responses to learn to rank them. RRHF only needs 1 to 2 models during tuning and can efficiently align language models with human preferences robustly without complex hyperparameter tuning. Additionally, RRHF can be considered an extension of SFT and reward model training while being simpler than PPO in terms of coding, model counts, and hyperparameters. We evaluate RRHF on the Helpful and Harmless dataset, demonstrating comparable alignment performance with PPO by reward model score and human labeling. Extensive experiments show that the performance of RRHF is highly related to sampling quality which suggests RRHF is a best-of-n learner. Codes available at https://github.com/GanjinZero/RRHF.

Zheng Yuan, Hongyi Yuan, Chuanqi Tan et al. · tsinghua

Large language models have emerged abilities including chain-of-thought to answer math word problems step by step. Solving math word problems not only requires abilities to disassemble problems via chain-of-thought but also needs to calculate arithmetic expressions correctly for each step. To the best of our knowledge, there is no work to focus on evaluating the arithmetic ability of large language models. In this work, we propose an arithmetic dataset MATH 401 to test the latest large language models including GPT-4, ChatGPT, InstrctGPT, Galactica, and LLaMA with various arithmetic expressions and provide a detailed analysis of the ability of large language models. MATH 401 and evaluation codes are released at \url{https://github.com/GanjinZero/math401-llm}.

Zheng Yuan, Chuanqi Tan, Songfang Huang · tsinghua

Automatic ICD coding is defined as assigning disease codes to electronic medical records (EMRs). Existing methods usually apply label attention with code representations to match related text snippets. Unlike these works that model the label with the code hierarchy or description, we argue that the code synonyms can provide more comprehensive knowledge based on the observation that the code expressions in EMRs vary from their descriptions in ICD. By aligning codes to concepts in UMLS, we collect synonyms of every code. Then, we propose a multiple synonyms matching network to leverage synonyms for better code representation learning, and finally help the code classification. Experiments on the MIMIC-III dataset show that our proposed method outperforms previous state-of-the-art methods.

Andrew Caines, Luca Benedetto, Shiva Taslimipoor et al. · cambridge

The recent release of very large language models such as PaLM and GPT-4 has made an unprecedented impact in the popular media and public consciousness, giving rise to a mixture of excitement and fear as to their capabilities and potential uses, and shining a light on natural language processing research which had not previously received so much attention. The developments offer great promise for education technology, and in this paper we look specifically at the potential for incorporating large language models in AI-driven language teaching and assessment systems. We consider several research areas and also discuss the risks and ethical considerations surrounding generative AI in education technology for language learners. Overall we find that larger language models offer improvements over previous models in text generation, opening up routes toward content generation which had not previously been plausible. For text generation they must be prompted carefully and their outputs may need to be reshaped before they are ready for use. For automated grading and grammatical error correction, tasks whose progress is checked on well-known benchmarks, early investigations indicate that large language models on their own do not improve on state-of-the-art results according to standard evaluation metrics. For grading it appears that linguistic features established in the literature should still be used for best performance, and for error correction it may be that the models can offer alternative feedback styles which are not measured sensitively with existing methods. In all cases, there is work to be done to experiment with the inclusion of large language models in education technology for language learners, in order to properly understand and report on their capacities and limitations, and to ensure that foreseeable risks such as misinformation and harmful bias are mitigated.

Hongyi Yuan, Zheng Yuan, Ruyi Gan et al. · tsinghua

Pretrained language models have served as important backbones for natural language processing. Recently, in-domain pretraining has been shown to benefit various domain-specific downstream tasks. In the biomedical domain, natural language generation (NLG) tasks are of critical importance, while understudied. Approaching natural language understanding (NLU) tasks as NLG achieves satisfying performance in the general domain through constrained language generation or language prompting. We emphasize the lack of in-domain generative language models and the unsystematic generative downstream benchmarks in the biomedical domain, hindering the development of the research community. In this work, we introduce the generative language model BioBART that adapts BART to the biomedical domain. We collate various biomedical language generation tasks including dialogue, summarization, entity linking, and named entity recognition. BioBART pretrained on PubMed abstracts has enhanced performance compared to BART and set strong baselines on several tasks. Furthermore, we conduct ablation studies on the pretraining tasks for BioBART and find that sentence permutation has negative effects on downstream tasks.

Christopher Bryant, Zheng Yuan, Muhammad Reza Qorib et al.

Grammatical Error Correction (GEC) is the task of automatically detecting and correcting errors in text. The task not only includes the correction of grammatical errors, such as missing prepositions and mismatched subject-verb agreement, but also orthographic and semantic errors, such as misspellings and word choice errors respectively. The field has seen significant progress in the last decade, motivated in part by a series of five shared tasks, which drove the development of rule-based methods, statistical classifiers, statistical machine translation, and finally neural machine translation systems which represent the current dominant state of the art. In this survey paper, we condense the field into a single article and first outline some of the linguistic challenges of the task, introduce the most popular datasets that are available to researchers (for both English and other languages), and summarise the various methods and techniques that have been developed with a particular focus on artificial error generation. We next describe the many different approaches to evaluation as well as concerns surrounding metric reliability, especially in relation to subjective human judgements, before concluding with an overview of recent progress and suggestions for future work and remaining challenges. We hope that this survey will serve as comprehensive resource for researchers who are new to the field or who want to be kept apprised of recent developments.

Sheng Yu, Zheng Yuan, Jun Xia et al. · tsinghua

Biomedical knowledge graphs (BioMedKGs) are essential infrastructures for biomedical and healthcare big data and artificial intelligence (AI), facilitating natural language processing, model development, and data exchange. For decades, these knowledge graphs have been developed via expert curation; however, this method can no longer keep up with today's AI development, and a transition to algorithmically generated BioMedKGs is necessary. In this work, we introduce the Biomedical Informatics Ontology System (BIOS), the first large-scale publicly available BioMedKG generated completely by machine learning algorithms. BIOS currently contains 4.1 million concepts, 7.4 million terms in two languages, and 7.3 million relation triplets. We present the methodology for developing BIOS, including the curation of raw biomedical terms, computational identification of synonymous terms and aggregation of these terms to create concept nodes, semantic type classification of the concepts, relation identification, and biomedical machine translation. We provide statistics on the current BIOS content and perform preliminary assessments of term quality, synonym grouping, and relation extraction. The results suggest that machine learning-based BioMedKG development is a viable alternative to traditional expert curation.

Anaïs Tack, Ekaterina Kochmar, Zheng Yuan et al.

This paper describes the results of the first shared task on the generation of teacher responses in educational dialogues. The goal of the task was to benchmark the ability of generative language models to act as AI teachers, replying to a student in a teacher-student dialogue. Eight teams participated in the competition hosted on CodaLab. They experimented with a wide variety of state-of-the-art models, including Alpaca, Bloom, DialoGPT, DistilGPT-2, Flan-T5, GPT-2, GPT-3, GPT- 4, LLaMA, OPT-2.7B, and T5-base. Their submissions were automatically scored using BERTScore and DialogRPT metrics, and the top three among them were further manually evaluated in terms of pedagogical ability based on Tack and Piech (2022). The NAISTeacher system, which ranked first in both automated and human evaluation, generated responses with GPT-3.5 using an ensemble of prompts and a DialogRPT-based ranking of responses for given dialogue contexts. Despite the promising achievements of the participating teams, the results also highlight the need for evaluation metrics better suited to educational contexts.

Stuart Mesham, Christopher Bryant, Marek Rei et al.

We extend a current sequence-tagging approach to Grammatical Error Correction (GEC) by introducing specialised tags for spelling correction and morphological inflection using the SymSpell and LemmInflect algorithms. Our approach improves generalisation: the proposed new tagset allows a smaller number of tags to correct a larger range of errors. Our results show a performance improvement both overall and in the targeted error categories. We further show that ensembles trained with our new tagset outperform those trained with the baseline tagset on the public BEA benchmark.

Chen Zhao, Wei-Ling Cai, Zheng Yuan et al.

Recently, unsupervised image-to-image translation methods based on contrastive learning have achieved state-of-the-art results in many tasks. However, in the previous works, the negatives are sampled from the input image itself, which inspires us to design a data augmentation method to improve the quality of the selected negatives. Moreover, the previous methods only preserve the content consistency via patch-wise contrastive learning in the embedding space, which ignores the domain consistency between the generated images and the real images of the target domain. In this paper, we propose a novel unsupervised image-to-image translation framework based on multi-cropping contrastive learning and domain consistency, called MCDUT. Specifically, we obtain the multi-cropping views via the center-cropping and the random-cropping with the aim of further generating the high-quality negative examples. To constrain the embeddings in the deep feature space, we formulate a new domain consistency loss, which encourages the generated images to be close to the real images in the embedding space of the same domain. Furthermore, we present a dual coordinate attention network by embedding positional information into the channel, which called DCA. We employ the DCA network in the design of generator, which makes the generator capture the horizontal and vertical global information of dependency. In many image-to-image translation tasks, our method achieves state-of-the-art results, and the advantages of our method have been proven through extensive comparison experiments and ablation research.

Christopher Davis, Andrew Caines, Øistein Andersen et al.

Thanks to recent advances in generative AI, we are able to prompt large language models (LLMs) to produce texts which are fluent and grammatical. In addition, it has been shown that we can elicit attempts at grammatical error correction (GEC) from LLMs when prompted with ungrammatical input sentences. We evaluate how well LLMs can perform at GEC by measuring their performance on established benchmark datasets. We go beyond previous studies, which only examined GPT* models on a selection of English GEC datasets, by evaluating seven open-source and three commercial LLMs on four established GEC benchmarks. We investigate model performance and report results against individual error types. Our results indicate that LLMs do not always outperform supervised English GEC models except in specific contexts -- namely commercial LLMs on benchmarks annotated with fluency corrections as opposed to minimal edits. We find that several open-source models outperform commercial ones on minimal edit benchmarks, and that in some settings zero-shot prompting is just as competitive as few-shot prompting.

Keren Tan, Kangyang Luo, Yunshi Lan et al.

Lexical Simplification (LS) aims to simplify text at the lexical level. Existing methods rely heavily on annotated data, making it challenging to apply in low-resource scenarios. In this paper, we propose a novel LS method without parallel corpora. This method employs an Adversarial Editing System with guidance from a confusion loss and an invariance loss to predict lexical edits in the original sentences. Meanwhile, we introduce an innovative LLM-enhanced loss to enable the distillation of knowledge from Large Language Models (LLMs) into a small-size LS system. From that, complex words within sentences are masked and a Difficulty-aware Filling module is crafted to replace masked positions with simpler words. At last, extensive experimental results and analyses on three benchmark LS datasets demonstrate the effectiveness of our proposed method.

Tom Potter, Zheng Yuan

With the rise of globalisation, code-switching (CSW) has become a ubiquitous part of multilingual conversation, posing new challenges for natural language processing (NLP), especially in Grammatical Error Correction (GEC). This work explores the complexities of applying GEC systems to CSW texts. Our objectives include evaluating the performance of state-of-the-art GEC systems on an authentic CSW dataset from English as a Second Language (ESL) learners, exploring synthetic data generation as a solution to data scarcity, and developing a model capable of correcting grammatical errors in monolingual and CSW texts. We generated synthetic CSW GEC data, resulting in one of the first substantial datasets for this task, and showed that a model trained on this data is capable of significant improvements over existing systems. This work targets ESL learners, aiming to provide educational technologies that aid in the development of their English grammatical correctness without constraining their natural multilingualism.

Kelvin Wey Han Chan, Christopher Bryant, Li Nguyen et al.

Code-switching (CSW) is a common phenomenon among multilingual speakers where multiple languages are used in a single discourse or utterance. Mixed language utterances may still contain grammatical errors however, yet most existing Grammar Error Correction (GEC) systems have been trained on monolingual data and not developed with CSW in mind. In this work, we conduct the first exploration into the use of GEC systems on CSW text. Through this exploration, we propose a novel method of generating synthetic CSW GEC datasets by translating different spans of text within existing GEC corpora. We then investigate different methods of selecting these spans based on CSW ratio, switch-point factor and linguistic constraints, and identify how they affect the performance of GEC systems on CSW text. Our best model achieves an average increase of 1.57 $F_{0.5}$ across 3 CSW test sets (English-Chinese, English-Korean and English-Japanese) without affecting the model's performance on a monolingual dataset. We furthermore discovered that models trained on one CSW language generalise relatively well to other typologically similar CSW languages.

Qiao Wang, Adnan Labib, Robert Swier et al.

GenQuest is a generative text adventure game that leverages Large Language Models (LLMs) to facilitate second language learning through immersive, interactive storytelling. The system engages English as a Foreign Language (EFL) learners in a collaborative "choose-your-own-adventure" style narrative, dynamically generated in response to learner choices. Game mechanics such as branching decision points and story milestones are incorporated to maintain narrative coherence while allowing learner-driven plot development. Key pedagogical features include content generation tailored to each learner's proficiency level, and a vocabulary assistant that provides in-context explanations of learner-queried text strings, ranging from words and phrases to sentences. Findings from a pilot study with university EFL students in China indicate promising vocabulary gains and positive user perceptions. Also discussed are suggestions from participants regarding the narrative length and quality, and the request for multi-modal content such as illustrations.

Yinsong Xu, Zheng Yuan

Inspired by Hosoyamada et al.'s work [14], we propose a new quantum meet-in-the-middle (QMITM) attack on $r$-round ($r \ge 7$) Feistel construction to reduce the time complexity. Similar to Hosoyamada et al.'s work, our attack on 7-round Feistel is also based on Guo et al.'s classical meet-in-the-middle (MITM) attack [13]. The classic MITM attack consumes a lot of time mainly in three aspects: construct the lookup table, query data and find a match. Therefore, parallel Grover search processors are used to reduce the time of constructing the lookup table. And we adjust the truncated differentials of the 5-round distinguisher proposed by Guo et al. to balance the complexities between constructing the lookup table and querying data. Finally, we introduce a quantum claw finding algorithm to find a match for reducing time. The subkeys can be recovered by this match. Furthermore, for $r$-round ($r > 7$) Feistel construction, we treat the above attack on the first 7 rounds as an inner loop and use Grover's algorithm to search the last $r-7$ rounds of subkeys as an outer loop. In summary, the total time complexity of our attack on $r$-round ($r \ge 7$) is only $O(2^{2n/3+(r-7)n/4})$ less than classical and quantum attacks. Moreover, our attack belongs to Q1 model and is more practical than other quantum attacks.

Ningyu Zhang, Mosha Chen, Zhen Bi et al.

Artificial Intelligence (AI), along with the recent progress in biomedical language understanding, is gradually changing medical practice. With the development of biomedical language understanding benchmarks, AI applications are widely used in the medical field. However, most benchmarks are limited to English, which makes it challenging to replicate many of the successes in English for other languages. To facilitate research in this direction, we collect real-world biomedical data and present the first Chinese Biomedical Language Understanding Evaluation (CBLUE) benchmark: a collection of natural language understanding tasks including named entity recognition, information extraction, clinical diagnosis normalization, single-sentence/sentence-pair classification, and an associated online platform for model evaluation, comparison, and analysis. To establish evaluation on these tasks, we report empirical results with the current 11 pre-trained Chinese models, and experimental results show that state-of-the-art neural models perform by far worse than the human ceiling. Our benchmark is released at \url{https://tianchi.aliyun.com/dataset/dataDetail?dataId=95414&lang=en-us}.

Marek Rei, Mariano Felice, Zheng Yuan et al.

Shortage of available training data is holding back progress in the area of automated error detection. This paper investigates two alternative methods for artificially generating writing errors, in order to create additional resources. We propose treating error generation as a machine translation task, where grammatically correct text is translated to contain errors. In addition, we explore a system for extracting textual patterns from an annotated corpus, which can then be used to insert errors into grammatically correct sentences. Our experiments show that the inclusion of artificially generated errors significantly improves error detection accuracy on both FCE and CoNLL 2014 datasets.