Jifan Yu

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.

Ji Qi, Jifan Yu, Teng Tu et al. · pku, tsinghua

Despite the recent emergence of video captioning models, how to generate vivid, fine-grained video descriptions based on the background knowledge (i.e., long and informative commentary about the domain-specific scenes with appropriate reasoning) is still far from being solved, which however has great applications such as automatic sports narrative. In this paper, we present GOAL, a benchmark of over 8.9k soccer video clips, 22k sentences, and 42k knowledge triples for proposing a challenging new task setting as Knowledge-grounded Video Captioning (KGVC). Moreover, we conduct experimental adaption of existing methods to show the difficulty and potential directions for solving this valuable and applicable task. Our data and code are available at https://github.com/THU-KEG/goal.

Jing Zhang, Xiaokang Zhang, Daniel Zhang-Li et al. · tsinghua

We present GLM-Dialog, a large-scale language model (LLM) with 10B parameters capable of knowledge-grounded conversation in Chinese using a search engine to access the Internet knowledge. GLM-Dialog offers a series of applicable techniques for exploiting various external knowledge including both helpful and noisy knowledge, enabling the creation of robust knowledge-grounded dialogue LLMs with limited proper datasets. To evaluate the GLM-Dialog more fairly, we also propose a novel evaluation method to allow humans to converse with multiple deployed bots simultaneously and compare their performance implicitly instead of explicitly rating using multidimensional metrics.Comprehensive evaluations from automatic to human perspective demonstrate the advantages of GLM-Dialog comparing with existing open source Chinese dialogue models. We release both the model checkpoint and source code, and also deploy it as a WeChat application to interact with users. We offer our evaluation platform online in an effort to prompt the development of open source models and reliable dialogue evaluation systems. The additional easy-to-use toolkit that consists of short text entity linking, query generation, and helpful knowledge classification is also released to enable diverse applications. All the source code is available on Github.

Shangqing Tu, Yuliang Sun, Yushi Bai et al. · tsinghua

To mitigate the potential misuse of large language models (LLMs), recent research has developed watermarking algorithms, which restrict the generation process to leave an invisible trace for watermark detection. Due to the two-stage nature of the task, most studies evaluate the generation and detection separately, thereby presenting a challenge in unbiased, thorough, and applicable evaluations. In this paper, we introduce WaterBench, the first comprehensive benchmark for LLM watermarks, in which we design three crucial factors: (1) For benchmarking procedure, to ensure an apples-to-apples comparison, we first adjust each watermarking method's hyper-parameter to reach the same watermarking strength, then jointly evaluate their generation and detection performance. (2) For task selection, we diversify the input and output length to form a five-category taxonomy, covering $9$ tasks. (3) For evaluation metric, we adopt the GPT4-Judge for automatically evaluating the decline of instruction-following abilities after watermarking. We evaluate $4$ open-source watermarks on $2$ LLMs under $2$ watermarking strengths and observe the common struggles for current methods on maintaining the generation quality. The code and data are available at https://github.com/THU-KEG/WaterBench.

Jifan Yu, Mengying Lu, Qingyang Zhong et al. · tsinghua

Student modeling, the task of inferring a student's learning characteristics through their interactions with coursework, is a fundamental issue in intelligent education. Although the recent attempts from knowledge tracing and cognitive diagnosis propose several promising directions for improving the usability and effectiveness of current models, the existing public datasets are still insufficient to meet the need for these potential solutions due to their ignorance of complete exercising contexts, fine-grained concepts, and cognitive labels. In this paper, we present MoocRadar, a fine-grained, multi-aspect knowledge repository consisting of 2,513 exercise questions, 5,600 knowledge concepts, and over 12 million behavioral records. Specifically, we propose a framework to guarantee a high-quality and comprehensive annotation of fine-grained concepts and cognitive labels. The statistical and experimental results indicate that our dataset provides the basis for the future improvements of existing methods. Moreover, to support the convenient usage for researchers, we release a set of tools for data querying, model adaption, and even the extension of our repository, which are now available at https://github.com/THU-KEG/MOOC-Radar.

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.

Zijun Yao, Yantao Liu, Xin Lv et al. · tsinghua

Deep text understanding, which requires the connections between a given document and prior knowledge beyond its text, has been highlighted by many benchmarks in recent years. However, these benchmarks have encountered two major limitations. On the one hand, most of them require human annotation of knowledge, which leads to limited knowledge coverage. On the other hand, they usually use choices or spans in the texts as the answers, which results in narrow answer space. To overcome these limitations, we build a new challenging benchmark named KoRc in this paper. Compared with previous benchmarks, KoRC has two advantages, i.e., broad knowledge coverage and flexible answer format. Specifically, we utilize massive knowledge bases to guide annotators or large language models (LLMs) to construct knowledgable questions. Moreover, we use labels in knowledge bases rather than spans or choices as the final answers. We test state-of-the-art models on KoRC and the experimental results show that the strongest baseline only achieves 68.3% and 30.0% F1 measure in the in-distribution and out-of-distribution test set, respectively. These results indicate that deep text understanding is still an unsolved challenge. The benchmark dataset, leaderboard, and baseline methods are released in https://github.com/THU-KEG/KoRC.

Shangqing Tu, Chunyang Li, Jifan Yu et al. · tsinghua

ChatGPT has achieved great success and can be considered to have acquired an infrastructural status. There are abundant works for evaluating ChatGPT on benchmarks. However, existing benchmarks encounter two challenges: (1) Disregard for periodical evaluation and (2) Lack of fine-grained features. In this paper, we construct ChatLog, an ever-updating dataset with large-scale records of diverse long-form ChatGPT responses for 21 NLP benchmarks from March, 2023 to now. We conduct a comprehensive performance evaluation to find that most capabilities of ChatGPT improve over time except for some abilities, and there exists a step-wise evolving pattern of ChatGPT. We further analyze the inherent characteristics of ChatGPT by extracting the knowledge and linguistic features. We find some stable features that stay unchanged and apply them on the detection of ChatGPT-generated texts to improve the robustness of cross-version detection. We will continuously maintain our project at \url{https://github.com/THU-KEG/ChatLog/}.

Sha Yuan, Hanyu Zhao, Shuai Zhao et al. · bytedance, pku

With the rapid development of deep learning, training Big Models (BMs) for multiple downstream tasks becomes a popular paradigm. Researchers have achieved various outcomes in the construction of BMs and the BM application in many fields. At present, there is a lack of research work that sorts out the overall progress of BMs and guides the follow-up research. In this paper, we cover not only the BM technologies themselves but also the prerequisites for BM training and applications with BMs, dividing the BM review into four parts: Resource, Models, Key Technologies and Application. We introduce 16 specific BM-related topics in those four parts, they are Data, Knowledge, Computing System, Parallel Training System, Language Model, Vision Model, Multi-modal Model, Theory&Interpretability, Commonsense Reasoning, Reliability&Security, Governance, Evaluation, Machine Translation, Text Generation, Dialogue and Protein Research. In each topic, we summarize clearly the current studies and propose some future research directions. At the end of this paper, we conclude the further development of BMs in a more general view.

Yushi Bai, Jiahao Ying, Yixin Cao et al. · tsinghua

Numerous benchmarks have been established to assess the performance of foundation models on open-ended question answering, which serves as a comprehensive test of a model's ability to understand and generate language in a manner similar to humans. Most of these works focus on proposing new datasets, however, we see two main issues within previous benchmarking pipelines, namely testing leakage and evaluation automation. In this paper, we propose a novel benchmarking framework, Language-Model-as-an-Examiner, where the LM serves as a knowledgeable examiner that formulates questions based on its knowledge and evaluates responses in a reference-free manner. Our framework allows for effortless extensibility as various LMs can be adopted as the examiner, and the questions can be constantly updated given more diverse trigger topics. For a more comprehensive and equitable evaluation, we devise three strategies: (1) We instruct the LM examiner to generate questions across a multitude of domains to probe for a broad acquisition, and raise follow-up questions to engage in a more in-depth assessment. (2) Upon evaluation, the examiner combines both scoring and ranking measurements, providing a reliable result as it aligns closely with human annotations. (3) We additionally propose a decentralized Peer-examination method to address the biases in a single examiner. Our data and benchmarking results are available at: http://lmexam.xlore.cn.

Bowen Zhao, Jiuding Sun, Bin Xu et al.

Web and artificial intelligence technologies, especially semantic web and knowledge graph (KG), have recently raised significant attention in educational scenarios. Nevertheless, subject-specific KGs for K-12 education still lack sufficiency and sustainability from knowledge and data perspectives. To tackle these issues, we propose EDUKG, a heterogeneous sustainable K-12 Educational Knowledge Graph. We first design an interdisciplinary and fine-grained ontology for uniformly modeling knowledge and resource in K-12 education, where we define 635 classes, 445 object properties, and 1314 datatype properties in total. Guided by this ontology, we propose a flexible methodology for interactively extracting factual knowledge from textbooks. Furthermore, we establish a general mechanism based on our proposed generalized entity linking system for EDUKG's sustainable maintenance, which can dynamically index numerous heterogeneous resources and data with knowledge topics in EDUKG. We further evaluate EDUKG to illustrate its sufficiency, richness, and variability. We publish EDUKG with more than 252 million entities and 3.86 billion triplets. Our code and data repository is now available at https://github.com/THU-KEG/EDUKG.

Daniel Zhang-Li, Zheyuan Zhang, Jifan Yu et al.

The vast pre-existing slides serve as rich and important materials to carry lecture knowledge. However, effectively leveraging lecture slides to serve students is difficult due to the multi-modal nature of slide content and the heterogeneous teaching actions. We study the problem of discovering effective designs that convert a slide into an interactive lecture. We develop Slide2Lecture, a tuning-free and knowledge-regulated intelligent tutoring system that can (1) effectively convert an input lecture slide into a structured teaching agenda consisting of a set of heterogeneous teaching actions; (2) create and manage an interactive lecture that generates responsive interactions catering to student learning demands while regulating the interactions to follow teaching actions. Slide2Lecture contains a complete pipeline for learners to obtain an interactive classroom experience to learn the slide. For teachers and developers, Slide2Lecture enables customization to cater to personalized demands. The evaluation rated by annotators and students shows that Slide2Lecture is effective in outperforming the remaining implementation. Slide2Lecture's online deployment has made more than 200K interaction with students in the 3K lecture sessions. We open source Slide2Lecture's implementation in https://anonymous.4open.science/r/slide2lecture-4210/.

Zijun Yao, Yuanyong Chen, Xin Lv et al. · tsinghua

We present Visual Knowledge oriented Programming platform (VisKoP), a knowledge base question answering (KBQA) system that integrates human into the loop to edit and debug the knowledge base (KB) queries. VisKoP not only provides a neural program induction module, which converts natural language questions into knowledge oriented program language (KoPL), but also maps KoPL programs into graphical elements. KoPL programs can be edited with simple graphical operators, such as dragging to add knowledge operators and slot filling to designate operator arguments. Moreover, VisKoP provides auto-completion for its knowledge base schema and users can easily debug the KoPL program by checking its intermediate results. To facilitate the practical KBQA on a million-entity-level KB, we design a highly efficient KoPL execution engine for the back-end. Experiment results show that VisKoP is highly efficient and user interaction can fix a large portion of wrong KoPL programs to acquire the correct answer. The VisKoP online demo https://demoviskop.xlore.cn (Stable release of this paper) and https://viskop.xlore.cn (Beta release with new features), highly efficient KoPL engine https://pypi.org/project/kopl-engine, and screencast video https://youtu.be/zAbJtxFPTXo are now publicly available.

Ji Qi, Kaixuan Ji, Xiaozhi Wang et al. · tsinghua

Open Information Extraction (OIE) aims to extract objective structured knowledge from natural texts, which has attracted growing attention to build dedicated models with human experience. As the large language models (LLMs) have exhibited remarkable in-context learning capabilities, a question arises as to whether the task of OIE can be effectively tackled with this paradigm? In this paper, we explore solving the OIE problem by constructing an appropriate reasoning environment for LLMs. Specifically, we first propose a method to effectively estimate the discrepancy of syntactic distribution between a LLM and test samples, which can serve as correlation evidence for preparing positive demonstrations. Upon the evidence, we introduce a simple yet effective mechanism to establish the reasoning environment for LLMs on specific tasks. Without bells and whistles, experimental results on the standard CaRB benchmark demonstrate that our $6$-shot approach outperforms state-of-the-art supervised method, achieving an $55.3$ $F_1$ score. Further experiments on TACRED and ACE05 show that our method can naturally generalize to other information extraction tasks, resulting in improvements of $5.7$ and $6.8$ $F_1$ scores, respectively.

Jinfeng Zhou, Zhuang Chen, Dazhen Wan et al.

In this paper, we present CharacterGLM, a series of models built upon ChatGLM, with model sizes ranging from 6B to 66B parameters. Our CharacterGLM is designed for generating Character-based Dialogues (CharacterDial), which aims to equip a conversational AI system with character customization for satisfying people's inherent social desires and emotional needs. On top of CharacterGLM, we can customize various AI characters or social agents by configuring their attributes (identities, interests, viewpoints, experiences, achievements, social relationships, etc.) and behaviors (linguistic features, emotional expressions, interaction patterns, etc.). Our model outperforms most mainstream close-source large langauge models, including the GPT series, especially in terms of consistency, human-likeness, and engagement according to manual evaluations. We will release our 6B version of CharacterGLM and a subset of training data to facilitate further research development in the direction of character-based dialogue generation.

Jifan Yu, Zheyuan Zhang, Daniel Zhang-li et al.

Since the first instances of online education, where courses were uploaded to accessible and shared online platforms, this form of scaling the dissemination of human knowledge to reach a broader audience has sparked extensive discussion and widespread adoption. Recognizing that personalized learning still holds significant potential for improvement, new AI technologies have been continuously integrated into this learning format, resulting in a variety of educational AI applications such as educational recommendation and intelligent tutoring. The emergence of intelligence in large language models (LLMs) has allowed for these educational enhancements to be built upon a unified foundational model, enabling deeper integration. In this context, we propose MAIC (Massive AI-empowered Course), a new form of online education that leverages LLM-driven multi-agent systems to construct an AI-augmented classroom, balancing scalability with adaptivity. Beyond exploring the conceptual framework and technical innovations, we conduct preliminary experiments at Tsinghua University, one of China's leading universities. Drawing from over 100,000 learning records of more than 500 students, we obtain a series of valuable observations and initial analyses. This project will continue to evolve, ultimately aiming to establish a comprehensive open platform that supports and unifies research, technology, and applications in exploring the possibilities of online education in the era of large model AI. We envision this platform as a collaborative hub, bringing together educators, researchers, and innovators to collectively explore the future of AI-driven online education.

Zheyuan Zhang, Jifan Yu, Juanzi Li et al.

Large Language Models (LLMs) have not only exhibited exceptional performance across various tasks, but also demonstrated sparks of intelligence. Recent studies have focused on assessing their capabilities on human exams and revealed their impressive competence in different domains. However, cognitive research on the overall knowledge structure of LLMs is still lacking. In this paper, based on educational diagnostic assessment method, we conduct an evaluation using MoocRadar, a meticulously annotated human test dataset based on Bloom Taxonomy. We aim to reveal the knowledge structures of LLMs and gain insights of their cognitive capabilities. This research emphasizes the significance of investigating LLMs' knowledge and understanding the disparate cognitive patterns of LLMs. By shedding light on models' knowledge, researchers can advance development and utilization of LLMs in a more informed and effective manner.

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.

Ji Qi, Bin Xu, Kaisheng Zeng et al.

Document-level relation extraction with graph neural networks faces a fundamental graph construction gap between training and inference - the golden graph structure only available during training, which causes that most methods adopt heuristic or syntactic rules to construct a prior graph as a pseudo proxy. In this paper, we propose $\textbf{ConstGCN}$, a novel graph convolutional network which performs knowledge-based information propagation between entities along with all specific relation spaces without any prior graph construction. Specifically, it updates the entity representation by aggregating information from all other entities along with each relation space, thus modeling the relation-aware spatial information. To control the information flow passing through the indeterminate relation spaces, we propose to constrain the propagation using transmitting scores learned from the Noise Contrastive Estimation between fact triples. Experimental results show that our method outperforms the previous state-of-the-art (SOTA) approaches on the DocRE dataset.

Xiaokang Zhang, Sijia Luo, Bohan Zhang et al. · tsinghua

We introduce TableLLM, a robust large language model (LLM) with 8 billion parameters, purpose-built for proficiently handling tabular data manipulation tasks, whether they are embedded within documents or spreadsheets, catering to real-world office scenarios. We propose a distant supervision method for training, which comprises a reasoning process extension strategy, aiding in training LLMs to understand reasoning patterns more effectively as well as a cross-way validation strategy, ensuring the quality of the automatically generated data. To evaluate the performance of TableLLM, we have crafted benchmarks tailored to address both document and spreadsheet formats as well as constructed a well-organized evaluation pipeline capable of handling both scenarios. Thorough evaluations underscore the advantages of TableLLM when compared to various existing general-purpose and tabular data-focused LLMs. We have publicly released the model checkpoint, source code, benchmarks, and a web application for user interaction. Our codes and data are publicly available at https://github.com/TableLLM/TableLLM.

Ji Qi, Kaixuan Ji, Jifan Yu et al.

Building models that comprehends videos and responds specific user instructions is a practical and challenging topic, as it requires mastery of both vision understanding and knowledge reasoning. Compared to language and image modalities, training efficiency remains a serious problem as existing studies train models on massive sparse videos paired with brief descriptions. In this paper, we introduce \textbf{VidCoM}, a fast adaptive framework that leverages Large Language Models (LLMs) to reason about videos using lightweight visual tools. Specifically, we reveal that the key to responding to specific instructions is focusing on relevant video events, and utilize two visual tools, structured scene graph generation and descriptive image caption generation, to gather and represent the event information. Thus, a LLM enriched with world knowledge is adopted as the reasoning agent to achieve the responses by performing multiple reasoning steps on specific video events. To address the difficulty of LLMs identifying video events, we further propose an Instruction-oriented Video Events Recognition (InsOVER) algorithm. This algorithm locates the corresponding video events based on an efficient Hungarian matching between decompositions of linguistic instructions and video events, thereby enabling LLMs to interact effectively with extended videos. Extensive experiments on two typical video comprehension tasks show that the proposed tuning-free framework outperforms the pre-trained models including Flamingo-80B, to achieve the state-of-the-art performance. Our source code and system will be publicly available.

Shangqing Tu, Yanjia Li, Keyu Chen et al.

Creating interactive STEM courseware traditionally requires HTML/CSS/JavaScript expertise, leaving barriers for educators. While generative AI can produce HTML codes, existing tools generate static presentations rather than interactive simulations, struggle with long documents, and lack pedagogical accuracy mechanisms. Furthermore, full regeneration for modifications requires 200--600 seconds, disrupting creative flow. We present MAIC-UI, a zero-code authoring system that enables educators to create and rapidly edit interactive courseware from textbooks, PPTs, and PDFs. MAIC-UI employs: (1) structured knowledge analysis with multi-modal understanding to ensure pedagogical rigor; (2) a two-stage generate-verify-optimize pipeline separating content alignment from visual refinement; and (3) Click-to-Locate editing with Unified Diff-based incremental generation achieving sub-10-second iteration cycles. A controlled lab study with 40 participants shows MAIC-UI reduces editing iterations (4.9 vs. 7.0) and significantly improves learnability and controllability compared to direct Text-to-HTML generation. A three-month classroom deployment with 53 high school students demonstrates that MAIC-UI fosters learning agency and reduces outcome disparities -- the pilot class achieved 9.21-point gains in STEM subjects compared to -2.32 points in control classes. Our code is available at https://github.com/THU-MAIC/MAIC-UI.

Fanjin Zhang, Kun Cao, Yukuo Cen et al. · tsinghua

Tracing the source of research papers is a fundamental yet challenging task for researchers. The billion-scale citation relations between papers hinder researchers from understanding the evolution of science efficiently. To date, there is still a lack of an accurate and scalable dataset constructed by professional researchers to identify the direct source of their studied papers, based on which automatic algorithms can be developed to expand the evolutionary knowledge of science. In this paper, we study the problem of paper source tracing (PST) and construct a high-quality and ever-increasing dataset PST-Bench in computer science. Based on PST-Bench, we reveal several intriguing discoveries, such as the differing evolution patterns across various topics. An exploration of various methods underscores the hardness of PST-Bench, pinpointing potential directions on this topic. The dataset and codes have been available at https://github.com/THUDM/paper-source-trace.

Zhanxin Hao, Jie Cao, Ruimiao Li et al.

Multi-agent AI systems, which simulate diverse instructional roles such as teachers and peers, offer new possibilities for personalized and interactive learning. Yet, student-AI interaction patterns and their pedagogical implications remain unclear. This study explores how university students engaged with multiple AI agents, and how these interactions influenced cognitive outcomes (learning gains) and non-cognitive factors (motivation, technology acceptance). Based on MAIC, an online learning platform with multi-agent, the research involved 305 university students and 19,365 lines of dialogue data. Pre- and post-test scores, self-reported motivation and technology acceptance were also collected. The study identified two engagement patterns: co-construction of knowledge and co-regulation. Lag sequential analysis revealed that students with lower prior knowledge relied more on co-construction of knowledge sequences, showing higher learning gains and post-course motivation. In contrast, students with higher prior knowledge engaged more in co-regulation behaviors but exhibited limited learning improvement. Technology acceptance increased across all groups. These findings suggest that multi-agent AI systems can adapt to students' varying needs, support differentiated engagement, and reduce performance gaps. Implications for personalized system design and future research directions are discussed.

Xiaokang Zhang, Zijun Yao, Jing Zhang et al.

Detecting non-factual content is a longstanding goal to increase the trustworthiness of large language models (LLMs) generations. Current factuality probes, trained using humanannotated labels, exhibit limited transferability to out-of-distribution content, while online selfconsistency checking imposes extensive computation burden due to the necessity of generating multiple outputs. This paper proposes PINOSE, which trains a probing model on offline self-consistency checking results, thereby circumventing the need for human-annotated data and achieving transferability across diverse data distributions. As the consistency check process is offline, PINOSE reduces the computational burden of generating multiple responses by online consistency verification. Additionally, it examines various aspects of internal states prior to response decoding, contributing to more effective detection of factual inaccuracies. Experiment results on both factuality detection and question answering benchmarks show that PINOSE achieves surpassing results than existing factuality detection methods. Our code and datasets are publicly available on this anonymized repository.

Bohan Zhang, Xiaokang Zhang, Jing Zhang et al.

Current inference scaling methods, such as Self-consistency and Best-of-N, have proven effective in improving the accuracy of LLMs on complex reasoning tasks. However, these methods rely heavily on the quality of candidate responses and are unable to produce correct answers when all candidates are incorrect. In this paper, we propose a novel inference scaling strategy, CoT-based Synthesizer, which leverages CoT reasoning to synthesize superior answers by analyzing complementary information from multiple candidate responses, even when all candidate responses are flawed. To enable a lightweight and cost-effective implementation, we introduce an automated data generation pipeline that creates diverse training data. This allows smaller LLMs trained on this data to improve the inference accuracy of larger models, including API-based LLMs. Experimental results across four benchmark datasets with seven policy models demonstrate that our method significantly enhances performance, with gains of 11.8% for Llama3-8B and 10.3% for GPT-4o on the MATH dataset. The corresponding training data and code are publicly available on https://github.com/RUCKBReasoning/CoT-based-Synthesizer.

Daniel Zhang-Li, Nianyi Lin, Jifan Yu et al.

Recent advancements in pretraining have demonstrated that modern Large Language Models (LLMs) possess the capability to effectively learn arithmetic operations. However, despite acknowledging the significance of digit order in arithmetic computation, current methodologies predominantly rely on sequential, step-by-step approaches for teaching LLMs arithmetic, resulting in a conclusion where obtaining better performance involves fine-grained step-by-step. Diverging from this conventional path, our work introduces a novel strategy that not only reevaluates the digit order by prioritizing output from the least significant digit but also incorporates a step-by-step methodology to substantially reduce complexity. We have developed and applied this method in a comprehensive set of experiments. Compared to the previous state-of-the-art (SOTA) method, our findings reveal an overall improvement of in accuracy while requiring only a third of the tokens typically used during training. For the purpose of facilitating replication and further research, we have made our code and dataset publicly available at \url{https://anonymous.4open.science/r/RAIT-9FB7/}.

Yuchen Fan, Yantao Liu, Zijun Yao et al.

Modern Large Language Models (LLMs) have showcased remarkable prowess in various tasks necessitating sophisticated cognitive behaviors. Nevertheless, a paradoxical performance discrepancy is observed, where these models underperform in seemingly elementary tasks like relation extraction and event extraction due to two issues in conventional evaluation. (1) The imprecision of existing evaluation metrics that struggle to effectively gauge semantic consistency between model outputs and ground truth, and (2) The inherent incompleteness of evaluation benchmarks, primarily due to restrictive human annotation schemas, resulting in underestimated LLM performances. Inspired by the principles in subjective question correction, we propose a new evaluation method, SQC-Score. This method innovatively utilizes LLMs, fine-tuned through subjective question correction data, to refine matching between model outputs and golden labels. Additionally, by incorporating a Natural Language Inference (NLI) model, SQC-Score enriches golden labels, addressing benchmark incompleteness by acknowledging correct yet previously omitted answers. Results on three information extraction tasks show that SQC-Score is more preferred by human annotators than the baseline metrics. Utilizing SQC-Score, we conduct a comprehensive evaluation of the state-of-the-art LLMs and provide insights for future research for information extraction. Dataset and associated codes can be accessed at https://github.com/THU-KEG/SQC-Score.

Shangqing Tu, Yucheng Wang, Daniel Zhang-Li et al. · tsinghua

Existing Large Vision-Language Models (LVLMs) can process inputs with context lengths up to 128k visual and text tokens, yet they struggle to generate coherent outputs beyond 1,000 words. We find that the primary limitation is the absence of long output examples during supervised fine-tuning (SFT). To tackle this issue, we introduce LongWriter-V-22k, a SFT dataset comprising 22,158 examples, each with multiple input images, an instruction, and corresponding outputs ranging from 0 to 10,000 words. Moreover, to achieve long outputs that maintain high-fidelity to the input images, we employ Direct Preference Optimization (DPO) to the SFT model. Given the high cost of collecting human feedback for lengthy outputs (e.g., 3,000 words), we propose IterDPO, which breaks long outputs into segments and uses iterative corrections to form preference pairs with the original outputs. Additionally, we develop MMLongBench-Write, a benchmark featuring six tasks to evaluate the long-generation capabilities of VLMs. Our 7B parameter model, trained with LongWriter-V-22k and IterDPO, achieves impressive performance on this benchmark, outperforming larger proprietary models like GPT-4o. Code and data: https://github.com/THU-KEG/LongWriter-V

Yantao Liu, Zijun Yao, Xin Lv et al.

Providing knowledge documents for large language models (LLMs) has emerged as a promising solution to update the static knowledge inherent in their parameters. However, knowledge in the document may conflict with the memory of LLMs due to outdated or incorrect knowledge in the LLMs' parameters. This leads to the necessity of examining the capability of LLMs to assimilate supplemental external knowledge that conflicts with their memory. While previous studies have explained to what extent LLMs extract conflicting knowledge from the provided text, they neglect the necessity to reason with conflicting knowledge. Furthermore, there lack a detailed analysis on strategies to enable LLMs to resolve conflicting knowledge via prompting, decoding strategy, and supervised fine-tuning. To address these limitations, we construct a new dataset, dubbed KNOT, for knowledge conflict resolution examination in the form of question answering. KNOT facilitates in-depth analysis by dividing reasoning with conflicting knowledge into three levels: (1) Direct Extraction, which directly extracts conflicting knowledge to answer questions. (2) Explicit Reasoning, which reasons with conflicting knowledge when the reasoning path is explicitly provided in the question. (3) Implicit Reasoning, where reasoning with conflicting knowledge requires LLMs to infer the reasoning path independently to answer questions. We also conduct extensive experiments on KNOT to establish empirical guidelines for LLMs to utilize conflicting knowledge in complex circumstances. Dataset and associated codes can be accessed at https://github.com/THU-KEG/KNOT .

Joy Lim Jia Yin, Daniel Zhang-Li, Jifan Yu et al.

Evaluating the quality of slide-based multimedia instruction is challenging. Existing methods like manual assessment, reference-based metrics, and large language model evaluators face limitations in scalability, context capture, or bias. In this paper, we introduce LecEval, an automated metric grounded in Mayer's Cognitive Theory of Multimedia Learning, to evaluate multimodal knowledge acquisition in slide-based learning. LecEval assesses effectiveness using four rubrics: Content Relevance (CR), Expressive Clarity (EC), Logical Structure (LS), and Audience Engagement (AE). We curate a large-scale dataset of over 2,000 slides from more than 50 online course videos, annotated with fine-grained human ratings across these rubrics. A model trained on this dataset demonstrates superior accuracy and adaptability compared to existing metrics, bridging the gap between automated and human assessments. We release our dataset and toolkits at https://github.com/JoylimJY/LecEval.

Joy Jia Yin Lim, Ye He, Jifan Yu et al.

Personalized Learning Path Planning (PLPP) aims to design adaptive learning paths that align with individual goals. While large language models (LLMs) show potential in personalizing learning experiences, existing approaches often lack mechanisms for goal-aligned planning. We introduce Pxplore, a novel framework for PLPP that integrates a reinforcement-based training paradigm and an LLM-driven educational architecture. We design a structured learner state model and an automated reward function that transforms abstract objectives into computable signals. We train the policy combining supervised fine-tuning (SFT) and Group Relative Policy Optimization (GRPO), and deploy it within a real-world learning platform. Extensive experiments validate Pxplore's effectiveness in producing coherent, personalized, and goal-driven learning paths. We release our code and dataset to facilitate future research.

Shangqing Tu, Zhuoran Pan, Wenxuan Wang et al.

Large language models (LLMs) have been increasingly applied to various domains, which triggers increasing concerns about LLMs' safety on specialized domains, e.g. medicine. Despite prior explorations on general jailbreaking attacks, there are two challenges for applying existing attacks on testing the domain-specific safety of LLMs: (1) Lack of professional knowledge-driven attacks, (2) Insufficient coverage of domain knowledge. To bridge this gap, we propose a new task, knowledge-to-jailbreak, which aims to generate jailbreaking attacks from domain knowledge, requiring both attack effectiveness and knowledge relevance. We collect a large-scale dataset with 12,974 knowledge-jailbreak pairs and fine-tune a large language model as jailbreak-generator, to produce domain knowledge-specific jailbreaks. Experiments on 13 domains and 8 target LLMs demonstrate the effectiveness of jailbreak-generator in generating jailbreaks that are both threatening to the target LLMs and relevant to the given knowledge. We also apply our method to an out-of-domain knowledge base, showing that jailbreak-generator can generate jailbreaks that are comparable in harmfulness to those crafted by human experts. Data and code are available at: https://github.com/THU-KEG/Knowledge-to-Jailbreak/.

Shangqing Tu, Yuanchun Wang, Jifan Yu et al.

Large language models have achieved remarkable success on general NLP tasks, but they may fall short for domain-specific problems. Recently, various Retrieval-Augmented Large Language Models (RALLMs) are proposed to address this shortcoming. However, existing evaluation tools only provide a few baselines and evaluate them on various domains without mining the depth of domain knowledge. In this paper, we address the challenges of evaluating RALLMs by introducing the R-Eval toolkit, a Python toolkit designed to streamline the evaluation of different RAG workflows in conjunction with LLMs. Our toolkit, which supports popular built-in RAG workflows and allows for the incorporation of customized testing data on the specific domain, is designed to be user-friendly, modular, and extensible. We conduct an evaluation of 21 RALLMs across three task levels and two representative domains, revealing significant variations in the effectiveness of RALLMs across different tasks and domains. Our analysis emphasizes the importance of considering both task and domain requirements when choosing a RAG workflow and LLM combination. We are committed to continuously maintaining our platform at https://github.com/THU-KEG/R-Eval to facilitate both the industry and the researchers.

Fangwei Zhu, Jifan Yu, Hailong Jin et al.

Entity linking models have achieved significant success via utilizing pretrained language models to capture semantic features. However, the NIL prediction problem, which aims to identify mentions without a corresponding entity in the knowledge base, has received insufficient attention. We categorize mentions linking to NIL into Missing Entity and Non-Entity Phrase, and propose an entity linking dataset NEL that focuses on the NIL prediction problem. NEL takes ambiguous entities as seeds, collects relevant mention context in the Wikipedia corpus, and ensures the presence of mentions linking to NIL by human annotation and entity masking. We conduct a series of experiments with the widely used bi-encoder and cross-encoder entity linking models, results show that both types of NIL mentions in training data have a significant influence on the accuracy of NIL prediction. Our code and dataset can be accessed at https://github.com/solitaryzero/NIL_EL

Ji Qi, Chuchun Zhang, Xiaozhi Wang et al.

The robustness to distribution changes ensures that NLP models can be successfully applied in the realistic world, especially for information extraction tasks. However, most prior evaluation benchmarks have been devoted to validating pairwise matching correctness, ignoring the crucial measurement of robustness. In this paper, we present the first benchmark that simulates the evaluation of open information extraction models in the real world, where the syntactic and expressive distributions under the same knowledge meaning may drift variously. We design and annotate a large-scale testbed in which each example is a knowledge-invariant clique that consists of sentences with structured knowledge of the same meaning but with different syntactic and expressive forms. By further elaborating the robustness metric, a model is judged to be robust if its performance is consistently accurate on the overall cliques. We perform experiments on typical models published in the last decade as well as a popular large language model, the results show that the existing successful models exhibit a frustrating degradation, with a maximum drop of 23.43 F1 score. Our resources and code are available at https://github.com/qijimrc/ROBUST.

Shulin Cao, Jiaxin Shi, Zijun Yao et al.

Program induction for answering complex questions over knowledge bases (KBs) aims to decompose a question into a multi-step program, whose execution against the KB produces the final answer. Learning to induce programs relies on a large number of parallel question-program pairs for the given KB. However, for most KBs, the gold program annotations are usually lacking, making learning difficult. In this paper, we propose the approach of program transfer, which aims to leverage the valuable program annotations on the rich-resourced KBs as external supervision signals to aid program induction for the low-resourced KBs that lack program annotations. For program transfer, we design a novel two-stage parsing framework with an efficient ontology-guided pruning strategy. First, a sketch parser translates the question into a high-level program sketch, which is the composition of functions. Second, given the question and sketch, an argument parser searches the detailed arguments from the KB for functions. During the searching, we incorporate the KB ontology to prune the search space. The experiments on ComplexWebQuestions and WebQuestionSP show that our method outperforms SOTA methods significantly, demonstrating the effectiveness of program transfer and our framework. Our codes and datasets can be obtained from https://github.com/THU-KEG/ProgramTransfer.

Zijun Yao, Chengjiang Li, Tiansi Dong et al.

Entity Matching (EM) aims at recognizing entity records that denote the same real-world object. Neural EM models learn vector representation of entity descriptions and match entities end-to-end. Though robust, these methods require many resources for training, and lack of interpretability. In this paper, we propose a novel EM framework that consists of Heterogeneous Information Fusion (HIF) and Key Attribute Tree (KAT) Induction to decouple feature representation from matching decision. Using self-supervised learning and mask mechanism in pre-trained language modeling, HIF learns the embeddings of noisy attribute values by inter-attribute attention with unlabeled data. Using a set of comparison features and a limited amount of annotated data, KAT Induction learns an efficient decision tree that can be interpreted by generating entity matching rules whose structure is advocated by domain experts. Experiments on 6 public datasets and 3 industrial datasets show that our method is highly efficient and outperforms SOTA EM models in most cases. Our codes and datasets can be obtained from https://github.com/THU-KEG/HIF-KAT.

Zhanxin Hao, Xiaobo Liu, Jiaxin Fan et al.

This study adopts an integrated distributed cognition and regulation of learning perspective to examine the collaboration patterns and dynamics of human-AI collaboration when college students collaborating with AI for complex problem-solving. Through cluster analysis, three distinct collaborative problem-solving modes were identified in this study: Delegated Reasoning (DR), Concerted Interpretation (CI), and Delegated Elaboration (DE). This study found that the DR group achieved the highest task performance, significantly outperforming the CI group. Additionally, the semantic similarity between human and AI discourse was notably the highest in the DR group. In contrast, the CI group reported significantly greater use of self-regulation strategies. These findings uncover a critical tension between the efficiency of the distributed system and the depth of human learners regulatory engagement. Insights from this study offer valuable implications for the future design of AI-empowered educational tools and student-AI collaborative learning frameworks.

Fanjin Zhang, Shijie Shi, Yifan Zhu et al. · tsinghua

With the rapid proliferation of scientific literature, versatile academic knowledge services increasingly rely on comprehensive academic graph mining. Despite the availability of public academic graphs, benchmarks, and datasets, these resources often fall short in multi-aspect and fine-grained annotations, are constrained to specific task types and domains, or lack underlying real academic graphs. In this paper, we present OAG-Bench, a comprehensive, multi-aspect, and fine-grained human-curated benchmark based on the Open Academic Graph (OAG). OAG-Bench covers 10 tasks, 20 datasets, 70+ baselines, and 120+ experimental results to date. We propose new data annotation strategies for certain tasks and offer a suite of data pre-processing codes, algorithm implementations, and standardized evaluation protocols to facilitate academic graph mining. Extensive experiments reveal that even advanced algorithms like large language models (LLMs) encounter difficulties in addressing key challenges in certain tasks, such as paper source tracing and scholar profiling. We also introduce the Open Academic Graph Challenge (OAG-Challenge) to encourage community input and sharing. We envisage that OAG-Bench can serve as a common ground for the community to evaluate and compare algorithms in academic graph mining, thereby accelerating algorithm development and advancement in this field. OAG-Bench is accessible at https://www.aminer.cn/data/.

Zeyao Ma, Xiaokang Zhang, Jing Zhang et al.

Current large language models (LLMs) often struggle to produce accurate responses on the first attempt for complex reasoning tasks like code generation. Prior research tackles this challenge by generating multiple candidate solutions and validating them with LLM-generated unit tests. The execution results of unit tests serve as reward signals to identify correct solutions. As LLMs always confidently make mistakes, these unit tests are not reliable, thereby diminishing the quality of reward signals. Motivated by the observation that scaling the number of solutions improves LLM performance, we explore the impact of scaling unit tests to enhance reward signal quality. Our pioneer experiment reveals a positive correlation between the number of unit tests and reward signal quality, with greater benefits observed in more challenging problems. Based on these insights, we propose CodeRM-8B, a lightweight yet effective unit test generator that enables efficient and high-quality unit test scaling. Additionally, we implement a dynamic scaling mechanism that adapts the number of unit tests based on problem difficulty, further improving efficiency. Experimental results show that our approach significantly improves performance across various models on three benchmarks (e.g., with gains of 18.43% for Llama3-8B and 3.42% for GPT-4o-mini on HumanEval Plus).

Haoxuan Li, Jifan Yu, Xin Cong et al.

Metacognitive education plays a crucial role in cultivating students' self-regulation and reflective thinking, providing essential support for those with learning difficulties through academic advising. Simulating students with insufficient learning capabilities using large language models offers a promising approach to refining pedagogical methods without ethical concerns. However, existing simulations often fail to authentically represent students' learning struggles and face challenges in evaluation due to the lack of reliable metrics and ethical constraints in data collection. To address these issues, we propose a pipeline for automatically generating and filtering high-quality simulated student agents. Our approach leverages a two-round automated scoring system validated by human experts and employs a score propagation module to obtain more consistent scores across the student graph. Experimental results demonstrate that our pipeline efficiently identifies high-quality student agents, and we discuss the traits that influence the simulation's effectiveness. By simulating students with varying degrees of learning difficulties, our work paves the way for broader applications in personalized learning and educational assessment.

Haoxuan Li, Jifan Yu, Yuanxin Ouyang et al.

Knowledge tracing (KT), aiming to mine students' mastery of knowledge by their exercise records and predict their performance on future test questions, is a critical task in educational assessment. While researchers achieved tremendous success with the rapid development of deep learning techniques, current knowledge tracing tasks fall into the cracks from real-world teaching scenarios. Relying heavily on extensive student data and solely predicting numerical performances differs from the settings where teachers assess students' knowledge state from limited practices and provide explanatory feedback. To fill this gap, we explore a new task formulation: Explainable Few-shot Knowledge Tracing. By leveraging the powerful reasoning and generation abilities of large language models (LLMs), we then propose a cognition-guided framework that can track the student knowledge from a few student records while providing natural language explanations. Experimental results from three widely used datasets show that LLMs can perform comparable or superior to competitive deep knowledge tracing methods. We also discuss potential directions and call for future improvements in relevant topics.

Jifan Yu, Xiaohan Zhang, Yifan Xu et al. · tsinghua

Empowered by the large-scale pretrained language models, existing dialogue systems have demonstrated impressive performance conducting fluent and natural-sounding conversations. However, they are still plagued by the hallucination problem, causing unpredictable factual errors in the generated responses. Recently, knowledge-grounded dialogue generation models, that intentionally invoke external knowledge resources to more informative responses, are also proven to be effective in reducing hallucination. Following the idea of getting high-quality knowledge, a few efforts have achieved pretty good performance on this issue. As some inevitable knowledge noises may also lead to hallucinations, it is emergent to investigate the reason and future directions for building noise-tolerant methods in KGD tasks. In this paper, we analyze the causal story behind this problem with counterfactual reasoning methods. Based on the causal effect analysis, we propose a possible solution for alleviating the hallucination in KGD by exploiting the dialogue-knowledge interaction. Experimental results of our example implementation show that this method can reduce hallucination without disrupting other dialogue performance, while keeping adaptive to different generation models. We hope our efforts can support and call for more attention to developing lightweight techniques towards robust and trusty dialogue systems.

Jie Cao, Zhanxin Hao, Jifan Yu

Educational dialogue is critical for decoding student learning processes, yet manual annotation remains time-consuming. This study evaluates the efficacy of GPT-5.2 and Gemini-3 using three prompting strategies (few-shot, single-agent, and multi-agent reflection) across diverse subjects, educational levels, and four coding dimensions. Results indicate that while multi-agent prompting achieved the highest accuracy, the results did not reach statistical significance. Accuracy proved highly context-dependent, with significantly higher performance in K-12 datasets compared to university-level data, alongside disciplinary variations within the same educational level. Performance peaked in the affective dimension but remained lowest in the cognitive dimension. Furthermore, analysis revealed four bias patterns: (1) Gemini-3 exhibited a consistent optimistic bias in the affective dimension across all subjects; (2) the cognitive dimension displayed domain-specific directional bias, characterized by systematic underestimation in Mathematics versus overestimation in Psychology; (3) both models are more prone to overestimation than underestimation within the meta-cognitive dimension; and (4) behavioral categories such as question, negotiation, and statements were frequently misclassified. These results underscore the need for context-sensitive deployment and targeted mitigation of directional biases in automated annotation.

Binglin Liu, Yucheng Wang, Zheyuan Zhang et al.

The adaptation of teaching slides to instructors' situated teaching needs, including pedagogical styles and their students' context, is a critical yet time-consuming task for educators. Through a series of educator interviews, we first identify and systematically categorize the key friction points that impede this adaptation process. Grounded in these findings, we introduce a novel multi-agent framework designed to automate slide adaptation based on high-level instructor specifications. An evaluation involving 16 modification requests across 8 real-world courses validates our approach. The framework's output consistently achieved high scores in intent alignment, content coherence and factual accuracy, and performed on par with baseline methods regarding visual clarity, while also demonstrating appropriate timeliness and a high operational agreement with human experts, achieving an F1 score of 0.89. This work heralds a new paradigm where AI agents handle the logistical burdens of instructional design, liberating educators to focus on the creative and strategic aspects of teaching.

Yuanchun Wang, Yiyang Fu, Jifan Yu et al.

Interactive online learning environments, represented by Massive AI-empowered Courses (MAIC), leverage LLM-driven multi-agent systems to transform passive MOOCs into dynamic, text-based platforms, enhancing interactivity through LLMs. This paper conducts an empirical study on a specific MAIC course to explore three research questions about dropouts in these interactive online courses: (1) What factors might lead to dropouts? (2) Can we predict dropouts? (3) Can we reduce dropouts? We analyze interaction logs to define dropouts and identify contributing factors. Our findings reveal strong links between dropout behaviors and textual interaction patterns. We then propose a course-progress-adaptive dropout prediction framework (CPADP) to predict dropouts with at most 95.4% accuracy. Based on this, we design a personalized email recall agent to re-engage at-risk students. Applied in the deployed MAIC system with over 3,000 students, the feasibility and effectiveness of our approach have been validated on students with diverse backgrounds.

Zheyuan Zhang, Daniel Zhang-Li, Jifan Yu et al.

Large language models (LLMs) have been applied across various intelligent educational tasks to assist teaching. While preliminary studies have focused on task-specific, independent LLM-empowered agents, the potential of LLMs within a multi-agent collaborative framework for classroom simulation with real user participation remains unexplored. In this work, we propose SimClass, a multi-agent classroom simulation teaching framework. We recognize representative class roles and introduce a novel class control mechanism for automatic classroom teaching, and conduct user experiments in two real-world courses. Using the Flanders Interactive Analysis System and Community of Inquiry theoretical frameworks from educational analysis, we demonstrate that LLMs can simulate a dynamic learning environment for users with active teacher-student and student-student interactions. We also observe group behaviors among agents in SimClass, where agents collaborate to create enlivening interactions in classrooms to improve user learning process. We hope this work pioneers the application of LLM-empowered multi-agent systems in virtual classroom teaching.

Zeyao Ma, Bohan Zhang, Jing Zhang et al.

We introduce SpreadsheetBench, a challenging spreadsheet manipulation benchmark exclusively derived from real-world scenarios, designed to immerse current large language models (LLMs) in the actual workflow of spreadsheet users. Unlike existing benchmarks that rely on synthesized queries and simplified spreadsheet files, SpreadsheetBench is built from 912 real questions gathered from online Excel forums, which reflect the intricate needs of users. The associated spreadsheets from the forums contain a variety of tabular data such as multiple tables, non-standard relational tables, and abundant non-textual elements. Furthermore, we propose a more reliable evaluation metric akin to online judge platforms, where multiple spreadsheet files are created as test cases for each instruction, ensuring the evaluation of robust solutions capable of handling spreadsheets with varying values. Our comprehensive evaluation of various LLMs under both single-round and multi-round inference settings reveals a substantial gap between the state-of-the-art (SOTA) models and human performance, highlighting the benchmark's difficulty.

Jing Zhang, Xiaokang Zhang, Jifan Yu et al.

Recent works on knowledge base question answering (KBQA) retrieve subgraphs for easier reasoning. A desired subgraph is crucial as a small one may exclude the answer but a large one might introduce more noises. However, the existing retrieval is either heuristic or interwoven with the reasoning, causing reasoning on the partial subgraphs, which increases the reasoning bias when the intermediate supervision is missing. This paper proposes a trainable subgraph retriever (SR) decoupled from the subsequent reasoning process, which enables a plug-and-play framework to enhance any subgraph-oriented KBQA model. Extensive experiments demonstrate SR achieves significantly better retrieval and QA performance than existing retrieval methods. Via weakly supervised pre-training as well as the end-to-end fine-tuning, SRl achieves new state-of-the-art performance when combined with NSM, a subgraph-oriented reasoner, for embedding-based KBQA methods.