Program-of-Thought (PoT) replaces natural language-based Chain-of-Thought (CoT) as the most popular method in Large Language Models (LLMs) mathematical reasoning tasks by utilizing external tool calls to circumvent computational errors. However, our evaluation of the GPT-4 and Llama series reveals that using PoT introduces more reasoning errors, such as incorrect formulas or flawed logic, compared to CoT. To address this issue, we propose Human-Think Language (HTL), which leverages a suite of strategies that help integrate PoT and CoT, encompassing: (1) a new generation paradigm that uses full CoT reasoning to control code generation. (2) Focus Attention, that directs model attention to the CoT reasoning during PoT to generate more logical code. (3) reinforcement learning that utilizes the accuracy of both CoT and PoT responses as rewards to prevent repetitive reasoning steps in LLMs when solving difficult math problems. Our method achieves an average improvement of 6.5% on the Llama-Base model and 4.3% on the Mistral-Base model across 8 mathematical calculation datasets. It also shows significant effectiveness on five out-of-domain datasets by controlling the model’s information flow, exhibiting strong transferability. Additionally, HTL shows the most significant improvement in non-mathematical natural language inference task, contributing to a unified reasoning task framework.

With the proliferation of Large Language Models (LLMs) in diverse domains, there is a particular need for unified evaluation standards in clinical medical scenarios, where models need to be examined very thoroughly. We present CliMedBench, a comprehensive benchmark with 14 expert-guided core clinical scenarios specifically designed to assess the medical ability of LLMs across 7 pivot dimensions. It comprises 33,735 questions derived from real-world medical reports of top-tier tertiary hospitals and authentic examination exercises. The reliability of this benchmark has been confirmed in several ways. Subsequent experiments with existing LLMs have led to the following findings: (i) Chinese medical LLMs underperform on this benchmark, especially where medical reasoning and factual consistency are vital, underscoring the need for advances in clinical knowledge and diagnostic accuracy. (ii) Several general-domain LLMs demonstrate substantial potential in medical clinics, while the limited input capacity of many medical LLMs hinders their practical use. These findings reveal both the strengths and limitations of LLMs in clinical scenarios and offer critical insights for medical research.

Domain adaptation from labeled source domains to the target domain is important in practical summarization scenarios. However, the key challenge is domain knowledge disentanglement. In this work, we explore how to disentangle domain-invariant knowledge from source domains while learning specific knowledge of the target domain. Specifically, we propose a hypernetwork-assisted encoder-decoder architecture with parameter-efficient fine-tuning. It leverages a hypernetwork instruction learning module to generate domain-specific parameters from the encoded inputs accompanied by task-related instruction. Further, to better disentangle and transfer knowledge from source domains to the target domain, we introduce a meta-knowledge distillation strategy to build a meta-teacher model that captures domain-invariant knowledge across multiple domains and use it to transfer knowledge to students. Experiments on three dialogue summarization datasets show the effectiveness of the proposed model. Human evaluations also show the superiority of our model with regard to the summary generation quality.

Chinese Spelling Correction (CSC) is the task of detecting and correcting misspelled charac- ters in Chinese texts. As an important step for various downstream tasks, CSC confronts two challenges: 1) Character-level errors consist not only of spelling errors but also of missing and redundant ones that cause variable length between input and output texts, for which most CSC methods could not handle well because of the consistence length of texts required by their inherent detection-correction framework. Con- sequently, the two errors are considered out- side the scope and left to future work, despite the fact that they are widely found and bound to CSC task in Chinese industrial scenario, such as Automatic Speech Recognition (ASR) and Optical Character Recognition (OCR). 2) Most existing CSC methods focus on either detector or corrector and train different mod- els for each one, respectively, leading to in- sufficiency of parameters sharing. To address these issues, we propose a novel model UMR- Spell to learn detection and correction parts together at the same time from a multi-task learning perspective by using a detection trans- mission self-attention matrix, and flexibly deal with both missing, redundant, and spelling er- rors through re-tagging rules. Furthermore, we build a new dataset ECMR-2023 containing five kinds of character-level errors to enrich the CSC task closer to real-world applications. Ex- periments on both SIGHAN benchmarks and ECMR-2023 demonstrate the significant effec- tiveness of UMRSpell over previous represen- tative baselines.

Succinctly summarizing dialogue is a task of growing interest, but inherent challenges, such as insufficient training data and low information density impede our ability to train abstractive models. In this work, we propose a novel curriculum-based prompt learning method with self-training to address these problems. Specifically, prompts are learned using a curriculum learning strategy that gradually increases the degree of prompt perturbation, thereby improving the dialogue understanding and modeling capabilities of our model. Unlabeled dialogue is incorporated by means of self-training so as to reduce the dependency on labeled data. We further investigate topic-aware prompts to better plan for the generation of summaries. Experiments confirm that our model substantially outperforms strong baselines and achieves new state-of-the-art results on the AMI and ICSI datasets. Human evaluations also show the superiority of our model with regard to the summary generation quality.

Generating explanations for recommender systems is essential for improving their transparency, as users often wish to understand the reason for receiving a specified recommendation. Previous methods mainly focus on improving the generation quality, but often produce generic explanations that fail to incorporate user and item specific details. To resolve this problem, we present Multi-Scale Distribution Deep Variational Autoencoders (MVAE).These are deep hierarchical VAEs with a prior network that eliminates noise while retaining meaningful signals in the input, coupled with a recognition network serving as the source of information to guide the learning of the prior network. Further, the Multi-scale distribution Learning Framework (MLF) along with a Target Tracking Kullback-Leibler divergence (TKL) mechanism are proposed to employ multi KL divergences at different scales for more effective learning. Extensive empirical experiments demonstrate that our methods can generate explanations with concrete input-specific contents.

This paper describes our system for SemEval-2021 Task 4: Reading Comprehension of Abstract Meaning. To accomplish this task, we utilize the Knowledge-Enhanced Graph Attention Network (KEGAT) architecture with a novel semantic space transformation strategy. It leverages heterogeneous knowledge to learn adequate evidences, and seeks for an effective semantic space of abstract concepts to better improve the ability of a machine in understanding the abstract meaning of natural language. Experimental results show that our system achieves strong performance on this task in terms of both imperceptibility and nonspecificity.

This paper describes our system for SemEval-2020 Task 4: Commonsense Validation and Explanation (Wang et al., 2020). We propose a novel Knowledge-enhanced Graph Attention Network (KEGAT) architecture for this task, leveraging heterogeneous knowledge from both the structured knowledge base (i.e. ConceptNet) and unstructured text to better improve the ability of a machine in commonsense understanding. This model has a powerful commonsense inference capability via utilizing suitable commonsense incorporation methods and upgraded data augmentation techniques. Besides, an internal sharing mechanism is cooperated to prohibit our model from insufficient and excessive reasoning for commonsense. As a result, this model performs quite well in both validation and explanation. For instance, it achieves state-of-the-art accuracy in the subtask called Commonsense Explanation (Multi-Choice). We officially name the system as ECNU-SenseMaker. Code is publicly available at https://github.com/ECNU-ICA/ECNU-SenseMaker.