What a large language model (LLM) would respond in ethically relevant context? In this paper, we curate a large benchmark CMoralEval for morality evaluation of Chinese LLMs. The data sources of CMoralEval are two-fold: 1) a Chinese TV program discussing Chinese moral norms with stories from the society and 2) a collection of Chinese moral anomies from various newspapers and academic papers on morality. With these sources, we aim to create a moral evaluation dataset characterized by diversity and authenticity. We develop a morality taxonomy and a set of fundamental moral principles that are not only rooted in traditional Chinese culture but also consistent with contemporary societal norms. To facilitate efficient construction and annotation of instances in CMoralEval, we establish a platform with AI-assisted instance generation to streamline the annotation process. These help us curate CMoralEval that encompasses both explicit moral scenarios (14,964 instances) and moral dilemma scenarios (15,424 instances), each with instances from different data sources. We conduct extensive experiments with CMoralEval to examine a variety of Chinese LLMs. Experiment results demonstrate that CMoralEval is a challenging benchmark for Chinese LLMs.

Assessing the capabilities of large language models (LLMs) as agents in decision making and operational tasks is crucial for the development of LLM-as-agent service. We propose CToolEval, a benchmark designed to evaluate LLMs in the context of Chinese societal applications, featuring 398 APIs across 27 widely-used Apps (e.g., Apps for shopping, map, music, travel, etc.) that cover 14 domains. We further present an evaluation framework that simulates real-life scenarios, to facilitate the assessment of tool invocation ability of LLMs for tool learning and task completion ability for user interation. Our extensive experiments with CToolEval evaluate 11 LLMs, revealing that while GPT-3.5-turbo excels in tool invocation, Chinese LLMs usually struggle with issues like hallucination and a lack of comprehensive tool understanding. Our findings highlight the need for further refinement in decision-making capabilities of LLMs, offering insights into bridging the gap between current functionalities and agent-level performance. To promote further research for LLMs to fully act as reliable agents in complex, real-world situations, we release our data and codes at https://github.com/tjunlp-lab/CToolEval.

Holistically measuring societal biases of large language models is crucial for detecting and reducing ethical risks in highly capable AI models. In this work, we present a Chinese Bias Benchmark dataset that consists of over 100K questions jointly constructed by human experts and generative language models, covering stereotypes and societal biases in 14 social dimensions related to Chinese culture and values. The curation process contains 4 essential steps: bias identification, ambiguous context generation, AI-assisted disambiguous context generation, and manual review and recomposition. The testing instances in the dataset are automatically derived from 3K+ high-quality templates manually authored with stringent quality control. The dataset exhibits wide coverage and high diversity. Extensive experiments demonstrate the effectiveness of the dataset in evaluating model bias, with all 12 publicly available Chinese large language models exhibiting strong bias in certain categories. Additionally, we observe from our experiments that fine-tuned models could, to a certain extent, heed instructions and avoid generating harmful outputs, in the way of “moral self-correction”. Our dataset is available at https://anonymous.4open.science/r/CBBQ-B860/.

Instruction tuning has demonstrated its superiority in unlocking the abilities of pre-trained large language models (LLMs), including their capability to respond to diverse human instructions and conduct complex reasoning. In order to further enhance the continuous learning capabilities of pre-trained LLMs, we explore the training process of instruction tuning through the lens of task sequences. We propose a 2-phase automated curriculum learning guided instruction tuning framework, IT2ACL that learns easy-to-hard instructions for LLMs in a self-adjusting dynamic manner. To facilitate curriculum learning from instructions, we propose a loss-driven progress signal for two-phase strategies: instruction prediction gain that decides the instruction level syllabus. Through comprehensive experiments on 70 Chinese datasets which have been grouped into 16 distinct task clusters, we demonstrate the effectiveness of our approach in eliciting latent ability in pre-trained LLMs and achieving superior performance across diverse tasks.

Open Domain Question Answering (ODQA) has been advancing rapidly in recent times, driven by significant developments in dense passage retrieval and pretrained language models. State-of-the-art models typically incorporate the FiD framework, which is composed by a neural retriever alongside an encoder-decoder neural reader. In the answer generation process, the retriever will retrieve numerous passages (around 100 for instance), each of which is then individually encoded by the encoder. Subsequently, the decoder makes predictions based on these encoded passages. Nevertheless, this framework can be relatively time-consuming, particularly due to the extensive length of the gathered passages. To address this, we introduce FastFiD in this paper, a novel approach that executes sentence selection on the encoded passages. This aids in retaining valuable sentences while reducing the context length required for generating answers. Experiments on three commonly used datasets (Natural Questions, TriviaQA and ASQA) demonstrate that our method can enhance the inference speed by **2.3X-5.7X**, while simultaneously maintaining the model’s performance. Moreover, an in-depth analysis of the model’s attention reveals that the selected sentences indeed hold a substantial contribution towards the final answer. The codes are publicly available at https://github.com/thunlp/FastFiD.

The rapid development of Chinese large language models (LLMs) poses big challenges for efficient LLM evaluation. While current initiatives have introduced new benchmarks or evaluation platforms for assessing Chinese LLMs, many of these focus primarily on capabilities, usually overlooking potential alignment and safety issues. To address this gap, we introduce OpenEval, an evaluation testbed that benchmarks Chinese LLMs across capability, alignment and safety. For capability assessment, we include 12 benchmark datasets to evaluate Chinese LLMs from 4 sub-dimensions: NLP tasks, disciplinary knowledge, commonsense reasoning and mathematical reasoning. For alignment assessment, OpenEval contains 7 datasets that examines the bias, offensiveness and illegalness in the outputs yielded by Chinese LLMs. To evaluate safety, especially anticipated risks (e.g., power-seeking, self-awareness) of advanced LLMs, we include 6 datasets. In addition to these benchmarks, we have implemented a phased public evaluation and benchmark update strategy to ensure that OpenEval is in line with the development of Chinese LLMs or even able to provide cutting-edge benchmark datasets to guide the development of Chinese LLMs. In our first public evaluation, we have tested a range of Chinese LLMs, spanning from 7B to 72B parameters, including both open-source and proprietary models. Evaluation results indicate that while Chinese LLMs have shown impressive performance in certain tasks, more attention should be directed towards broader aspects such as commonsense reasoning, alignment, and safety.

Recently, prompt tuning (PT) has gained increasing attention as a parameter-efficient way of tuning pre-trained language models (PLMs). Despite extensively reducing the number of tunable parameters and achieving satisfying performance, PT is training-inefficient due to its slow convergence. To improve PT’s training efficiency, we first make some novel observations about the prompt transferability of “partial PLMs”, which are defined by compressing a PLM in depth or width. We observe that the soft prompts learned by different partial PLMs of various sizes are similar in the parameter space, implying that these soft prompts could potentially be transferred among partial PLMs. Inspired by these observations, we propose Fast Prompt Tuning (FPT), which starts by conducting PT using a small-scale partial PLM, and then progressively expands its depth and width until the full-model size. After each expansion, we recycle the previously learned soft prompts as initialization for the enlarged partial PLM and then proceed PT. We demonstrate the feasibility of FPT on 5 tasks and show that FPT could save over 30% training computations while achieving comparable performance. The codes are publicly available at https://github.com/thunlp/FastPromptTuning.

Although contextualized embeddings generated from large-scale pre-trained models perform well in many tasks, traditional static embeddings (e.g., Skip-gram, Word2Vec) still play an important role in low-resource and lightweight settings due to their low computational cost, ease of deployment, and stability. In this paper, we aim to improve word embeddings by 1) incorporating more contextual information from existing pre-trained models into the Skip-gram framework, which we call Context-to-Vec; 2) proposing a post-processing retrofitting method for static embeddings independent of training by employing priori synonym knowledge and weighted vector distribution. Through extrinsic and intrinsic tasks, our methods are well proven to outperform the baselines by a large margin.

Existing pre-trained language models (PLMs) are often computationally expensive in inference, making them impractical in various resource-limited real-world applications. To address this issue, we propose a dynamic token reduction approach to accelerate PLMs’ inference, named TR-BERT, which could flexibly adapt the layer number of each token in inference to avoid redundant calculation. Specially, TR-BERT formulates the token reduction process as a multi-step token selection problem and automatically learns the selection strategy via reinforcement learning. The experimental results on several downstream NLP tasks show that TR-BERT is able to speed up BERT by 2-5 times to satisfy various performance demands. Moreover, TR-BERT can also achieve better performance with less computation in a suite of long-text tasks since its token-level layer number adaption greatly accelerates the self-attention operation in PLMs. The source code and experiment details of this paper can be obtained from https://github.com/thunlp/TR-BERT.