Zheyuan Zhang

Jifan Yu, Xiaozhi Wang, Shangqing Tu et al. · tsinghua

The unprecedented performance of large language models (LLMs) necessitates improvements in evaluations. Rather than merely exploring the breadth of LLM abilities, we believe meticulous and thoughtful designs are essential to thorough, unbiased, and applicable evaluations. Given the importance of world knowledge to LLMs, we construct a Knowledge-oriented LLM Assessment benchmark (KoLA), in which we carefully design three crucial factors: (1) For \textbf{ability modeling}, we mimic human cognition to form a four-level taxonomy of knowledge-related abilities, covering $19$ tasks. (2) For \textbf{data}, to ensure fair comparisons, we use both Wikipedia, a corpus prevalently pre-trained by LLMs, along with continuously collected emerging corpora, aiming to evaluate the capacity to handle unseen data and evolving knowledge. (3) For \textbf{evaluation criteria}, we adopt a contrastive system, including overall standard scores for better numerical comparability across tasks and models and a unique self-contrast metric for automatically evaluating knowledge-creating ability. We evaluate $28$ open-source and commercial LLMs and obtain some intriguing findings. The KoLA dataset and open-participation leaderboard are publicly released at https://kola.xlore.cn and will be continuously updated to provide references for developing LLMs and knowledge-related systems.

Shangqing Tu, Zheyuan Zhang, Jifan Yu et al. · tsinghua

Teaching assistants have played essential roles in the long history of education. However, few MOOC platforms are providing human or virtual teaching assistants to support learning for massive online students due to the complexity of real-world online education scenarios and the lack of training data. In this paper, we present a virtual MOOC teaching assistant, LittleMu with minimum labeled training data, to provide question answering and chit-chat services. Consisting of two interactive modules of heterogeneous retrieval and language model prompting, LittleMu first integrates structural, semi- and unstructured knowledge sources to support accurate answers for a wide range of questions. Then, we design delicate demonstrations named "Chain of Teach" prompts to exploit the large-scale pre-trained model to handle complex uncollected questions. Except for question answering, we develop other educational services such as knowledge-grounded chit-chat. We test the system's performance via both offline evaluation and online deployment. Since May 2020, our LittleMu system has served over 80,000 users with over 300,000 queries from over 500 courses on XuetangX MOOC platform, which continuously contributes to a more convenient and fair education. Our code, services, and dataset will be available at https://github.com/THU-KEG/VTA.

Zheyuan Zhang, Lanhong Yao, Bin Wang et al.

Large-scale, big-variant, high-quality data are crucial for developing robust and successful deep-learning models for medical applications since they potentially enable better generalization performance and avoid overfitting. However, the scarcity of high-quality labeled data always presents significant challenges. This paper proposes a novel approach to address this challenge by developing controllable diffusion models for medical image synthesis, called DiffBoost. We leverage recent diffusion probabilistic models to generate realistic and diverse synthetic medical image data that preserve the essential characteristics of the original medical images by incorporating edge information of objects to guide the synthesis process. In our approach, we ensure that the synthesized samples adhere to medically relevant constraints and preserve the underlying structure of imaging data. Due to the random sampling process by the diffusion model, we can generate an arbitrary number of synthetic images with diverse appearances. To validate the effectiveness of our proposed method, we conduct an extensive set of medical image segmentation experiments on multiple datasets, including Ultrasound breast (+13.87%), CT spleen (+0.38%), and MRI prostate (+7.78%), achieving significant improvements over the baseline segmentation methods. The promising results demonstrate the effectiveness of our \textcolor{black}{DiffBoost} for medical image segmentation tasks and show the feasibility of introducing a first-ever text-guided diffusion model for general medical image segmentation tasks. With carefully designed ablation experiments, we investigate the influence of various data augmentations, hyper-parameter settings, patch size for generating random merging mask settings, and combined influence with different network architectures. Source code are available at https://github.com/NUBagciLab/DiffBoost.

Bin Wang, Armstrong Aboah, Zheyuan Zhang et al.

This study investigates the potential of eye-tracking technology and the Segment Anything Model (SAM) to design a collaborative human-computer interaction system that automates medical image segmentation. We present the \textbf{GazeSAM} system to enable radiologists to collect segmentation masks by simply looking at the region of interest during image diagnosis. The proposed system tracks radiologists' eye movement and utilizes the eye-gaze data as the input prompt for SAM, which automatically generates the segmentation mask in real time. This study is the first work to leverage the power of eye-tracking technology and SAM to enhance the efficiency of daily clinical practice. Moreover, eye-gaze data coupled with image and corresponding segmentation labels can be easily recorded for further advanced eye-tracking research. The code is available in \url{https://github.com/ukaukaaaa/GazeSAM}.

Qingyang Zhong, Jifan Yu, Zheyuan Zhang et al.

Adaptive learning aims to stimulate and meet the needs of individual learners, which requires sophisticated system-level coordination of diverse tasks, including modeling learning resources, estimating student states, and making personalized recommendations. Existing deep learning methods have achieved great success over statistical models; however, they still lack generalization for diverse tasks and suffer from insufficient capacity since they are composed of highly-coupled task-specific architectures and rely on small-scale, coarse-grained recommendation scenarios. To realize the idea of general adaptive systems proposed in pedagogical theory, with the emerging pre-training techniques in NLP, we try to conduct a practical exploration on applying pre-training to adaptive learning, to propose a unified framework based on data observation and learning style analysis, properly leveraging heterogeneous learning elements. Through a series of downstream tasks of Learning Recommendation, Learning Resource Evaluation, Knowledge Tracing, and Dropout Prediction, we find that course structures, text, and knowledge are helpful for modeling and inherently coherent to student non-sequential learning behaviors and that indirectly relevant information included in the pre-training foundation can be shared across downstream tasks to facilitate effectiveness. We finally build a simplified systematic application of adaptive learning and reflect on the insights brought back to pedagogy. The source code and dataset will be released.

Zheyuan Zhang, Shane Storks, Fengyuan Hu et al.

Pre-trained language models (PLMs) have shown impressive performance in various language tasks. However, they are prone to spurious correlations, and often generate illusory information. In real-world applications, PLMs should justify decisions with formalized, coherent reasoning chains, but this challenge remains under-explored. Cognitive psychology theorizes that humans are capable of utilizing fast and intuitive heuristic thinking to make decisions based on past experience, then rationalizing the decisions through slower and deliberative analytic reasoning. We incorporate these interlinked dual processes in fine-tuning and in-context learning with PLMs, applying them to two language understanding tasks that require coherent physical commonsense reasoning. We show that our proposed Heuristic-Analytic Reasoning (HAR) strategies drastically improve the coherence of rationalizations for model decisions, yielding state-of-the-art results on Tiered Reasoning for Intuitive Physics (TRIP). We also find that this improved coherence is a direct result of more faithful attention to relevant language context in each step of reasoning. Our findings suggest that human-like reasoning strategies can effectively improve the coherence and reliability of PLM reasoning.

Chris Sanchez, Zheyuan Zhang

Many prior language modeling efforts have shown that pre-training on an in-domain corpus can significantly improve performance on downstream domain-specific NLP tasks. However, the difficulties associated with collecting enough in-domain data might discourage researchers from approaching this pre-training task. In this paper, we conducted a series of experiments by pre-training Bidirectional Encoder Representations from Transformers (BERT) with different sizes of biomedical corpora. The results demonstrate that pre-training on a relatively small amount of in-domain data (4GB) with limited training steps, can lead to better performance on downstream domain-specific NLP tasks compared with fine-tuning models pre-trained on general corpora.

Zheyuan Zhang, Elif Keles, Gorkem Durak et al.

Automated volumetric segmentation of the pancreas on cross-sectional imaging is needed for diagnosis and follow-up of pancreatic diseases. While CT-based pancreatic segmentation is more established, MRI-based segmentation methods are understudied, largely due to a lack of publicly available datasets, benchmarking research efforts, and domain-specific deep learning methods. In this retrospective study, we collected a large dataset (767 scans from 499 participants) of T1-weighted (T1W) and T2-weighted (T2W) abdominal MRI series from five centers between March 2004 and November 2022. We also collected CT scans of 1,350 patients from publicly available sources for benchmarking purposes. We developed a new pancreas segmentation method, called PanSegNet, combining the strengths of nnUNet and a Transformer network with a new linear attention module enabling volumetric computation. We tested PanSegNet's accuracy in cross-modality (a total of 2,117 scans) and cross-center settings with Dice and Hausdorff distance (HD95) evaluation metrics. We used Cohen's kappa statistics for intra and inter-rater agreement evaluation and paired t-tests for volume and Dice comparisons, respectively. For segmentation accuracy, we achieved Dice coefficients of 88.3% (std: 7.2%, at case level) with CT, 85.0% (std: 7.9%) with T1W MRI, and 86.3% (std: 6.4%) with T2W MRI. There was a high correlation for pancreas volume prediction with R^2 of 0.91, 0.84, and 0.85 for CT, T1W, and T2W, respectively. We found moderate inter-observer (0.624 and 0.638 for T1W and T2W MRI, respectively) and high intra-observer agreement scores. All MRI data is made available at https://osf.io/kysnj/. Our source code is available at https://github.com/NUBagciLab/PaNSegNet.

Shane Storks, Itamar Bar-Yossef, Yayuan Li et al.

Procedural mistake detection (PMD) is a challenging problem of classifying whether a human user (observed through egocentric video) has successfully executed a task (specified by a procedural text). Despite significant recent efforts, machine performance in the wild remains nonviable, and the reasoning processes underlying this performance are opaque. As such, we extend PMD to require generating visual self-dialog rationales to inform decisions. Given the impressive, mature image understanding capabilities observed in recent vision-and-language models (VLMs), we curate a suitable benchmark dataset for PMD based on individual frames. As our reformulation enables unprecedented transparency, we leverage a natural language inference (NLI) model to formulate two automated metrics for the coherence of generated rationales. We establish baselines for this reframed task, showing that VLMs struggle off-the-shelf, but with some trade-offs, their accuracy, coherence, and efficiency can be improved by incorporating these metrics into common inference and fine-tuning methods. Lastly, our multi-faceted metrics visualize common outcomes, highlighting areas for further improvement.