This paper describes our approaches for the BEA-2025 Shared Task on assessing pedagogical ability and attributing tutor identities in AI-powered tutoring systems. We explored three methodological paradigms: in-context learning (ICL), supervised fine-tuning (SFT), and reinforcement learning from human feedback (RLHF). Results indicate clear methodological strengths: SFT is highly effective for structured classification tasks such as mistake identification and feedback actionability, while ICL with advanced prompting excels at open-ended tasks involving mistake localization and instructional guidance. Additionally, fine-tuned models demonstrated strong performance in identifying tutor authorship. Our findings highlight the importance of aligning methodological strategy and task structure, providing insights toward more effective evaluations of educational AI systems.

Entity Alignment (EA) is a critical task in Knowledge Graph (KG) integration, aimed at identifying and matching equivalent entities that represent the same real-world objects. While EA methods based on knowledge representation learning have shown strong performance on synthetic benchmark datasets such as DBP15K, their effectiveness significantly decline in real-world scenarios which often involve data that is highly heterogeneous, incomplete, and domain-specific, as seen in datasets like DOREMUS and AGROLD. Addressing this challenge, we propose DAEA, a novel EA approach with Domain Adaptation that leverages the data characteristics of synthetic benchmarks for improved performance in real-world datasets. DAEA introduces a multi-source KGs selection mechanism and a specialized domain adaptive entity alignment loss function to bridge the gap between real-world data and optimal benchmark data, mitigating the challenges posed by aligning entities across highly heterogeneous KGs. Experimental results demonstrate that DAEA outperforms state-of-the-art models on real-world datasets, achieving a 29.94% improvement in Hits@1 on DOREMUS and a 5.64% improvement on AGROLD. Code is available at https://github.com/yangxiaoxiaoly/DAEA.

With the different roles that AI is expected to play in human life, imbuing large language models (LLMs) with different personalities has attracted increasing research interest. While the “personification” enhances human experiences of interactivity and adaptability of LLMs, it gives rise to critical concerns about content safety, particularly regarding bias, sentiment, and toxicity of LLM generation. This study explores how assigning different personality traits to LLMs affects the toxicity and biases of their outputs. Leveraging the widely accepted HEXACO personality framework developed in social psychology, we design experimentally sound prompts to test three LLMs’ performance on three toxic and bias benchmarks. The findings demonstrate the sensitivity of all three models to HEXACO personality traits and, more importantly, a consistent variation in the biases, negative sentiment, and toxicity of their output. In particular, adjusting the levels of several personality traits can effectively reduce bias and toxicity in model performance, similar to humans’ correlations between personality traits and toxic behaviors. The findings highlight the additional need to examine content safety besides the efficiency of training or fine-tuning methods for LLM personification, they also suggest a potential for the adjustment of personalities to be a simple and low-cost method to conduct controlled text generation.

The rapid development of large language models (LLMs) gives rise to ethical concerns about their performance, while opening new avenues for developing toxic language detection techniques. However, LLMs’ unethical output and their capability of detecting toxicity have primarily been tested on language data that do not demand complex meaning inference, such as the biased associations of ‘he’ with programmer and ‘she’ with household. Nowadays toxic language adopts a much more creative range of implicit forms, thanks to advanced censorship. In this study, we collect authentic toxic interactions that evade online censorship and that are verified by human annotators as inference-intensive. To evaluate and improve LLMs’ reasoning of the authentic implicit toxic language, we propose a new prompting method, Pragmatic Inference Chain (PIC), drawn on interdisciplinary findings from cognitive science and linguistics. The PIC prompting significantly improves the success rate of GPT-4o, Llama-3.1-70B-Instruct, DeepSeek-v2.5, and DeepSeek-v3 in identifying implicit toxic language, compared to five baseline prompts, such as CoT and rule-based baselines. In addition, it also facilitates the models to produce more explicit and coherent reasoning processes, hence can potentially be generalized to other inference-intensive tasks, e.g., understanding humour and metaphors.

Large Language Models (LLMs) fine-tuning technologies have achieved remarkable results. However, traditional LLM fine-tuning approaches face significant challenges: they require large Floating Point(FP) computation, raising privacy concerns when handling sensitive data, and are impractical for resource-constrained edge devices. While Parameter-Efficient Fine-Tuning (PEFT) techniques reduce trainable parameters, their reliance on floating-point arithmetic creates fundamental incompatibilities with edge hardware. In this work, we introduce a novel framework for on-device LLM fine-tuning that eliminates the need for floating-point operations in both inference and training, named GSQ-Tuning. At its core is the Group-Shared Exponents Integer format, which efficiently represents model parameters in integer format using shared exponents among parameter groups. When combined with LoRA-like adapters, this enables fully integer-based fine-tuning that is both memory and compute efficient. We demonstrate that our approach achieves accuracy comparable to FP16-based fine-tuning while significantly reducing memory usage ( 50%). Moreover, compared to FP8, at comparable performance levels, our method can reduce 5x power consumption and 11x chip area, making large-scale model adaptation feasible on edge devices.

While large language models (LLMs) have been pre-trained on multilingual corpora, their performance still lags behind in most languages compared to a few resource-rich languages. One common approach to mitigate this issue is to translate training data from resource-rich languages into other languages and then continue training. However, using the data obtained solely relying on translation while ignoring the original capabilities of LLMs across languages is not always effective, which we show will limit the performance of cross-lingual knowledge transfer. In this work, we propose SDRRL, a method based on Self-Distillation from Resource-Rich Languages that effectively improve multilingual performance by leveraging the internal capabilities of LLMs on resource-rich languages. We evaluate on different LLMs (LLaMA-2 and SeaLLM) and source languages (English and French) across various comprehension and generation tasks, experimental results demonstrate that SDRRL can significantly enhance multilingual capabilities while minimizing the impact on original performance in resource-rich languages.

Open-source large language models (LLMs) have gained significant strength across diverse fields. Nevertheless, the majority of studies primarily concentrate on English, with only limited exploration into the realm of multilingual abilities.In this work, we therefore construct an open-source multilingual supervised fine-tuning dataset.Different from previous works that simply translate English instructions, we consider both the language-specific and language-agnostic abilities of LLMs. Firstly, we introduce a knowledge-grounded data augmentation approach to elicit more language-specific knowledge of LLMs, improving their ability to serve users from different countries. Moreover, we find modern LLMs possess strong cross-lingual transfer capabilities, thus repeatedly learning identical content in various languages is not necessary. Consequently, we can substantially prune the language-agnostic supervised fine-tuning (SFT) data without any performance degradation, making multilingual SFT more efficient.The resulting UltraLink dataset comprises approximately 1 million samples across five languages (i.e., En, Zh, Ru, Fr, Es), and the proposed data construction method can be easily extended to other languages.UltraLink-LM, which is trained on the UltraLink dataset, outperforms several representative baselines across many tasks.

LoRA employs lightweight modules to customize large language models (LLMs) for each downstream task or domain, where different learned additional modules represent diverse skills. Combining existing LoRAs to address new tasks can enhance the reusability of learned LoRAs, particularly beneficial for tasks with limited annotated data. Most prior works on LoRA combination primarily rely on task-level weights for each involved LoRA, making different examples and tokens share the same LoRA weights. However, in generative tasks, different tokens may necessitate diverse skills to manage. Taking the Chinese math task as an example, understanding the problem description may depend more on the Chinese LoRA, while the calculation part may rely more on the math LoRA. To this end, we propose LoRA-Flow, which utilizes dynamic weights to adjust the impact of different LoRAs. The weights at each step are determined by a fusion gate with extremely few parameters, which can be learned with only 200 training examples. Experiments across six generative tasks demonstrate that our method consistently outperforms baselines with task-level fusion weights. This underscores the necessity of introducing dynamic fusion weights for LoRA combination.

Processing and reasoning over long contexts is crucial for many practical applications of Large Language Models (LLMs), such as document comprehension and agent construction. Despite recent strides in making LLMs process contexts with more than 100K tokens, there is currently a lack of a standardized benchmark to evaluate this long-context capability. Existing public benchmarks typically focus on contexts around 10K tokens, limiting the assessment and comparison of LLMs in processing longer contexts. In this paper, we propose , the first LLM benchmark featuring an average data length surpassing 100K tokens. comprises synthetic and realistic tasks spanning diverse domains in English and Chinese. The tasks in are designed to require an understanding of long dependencies in contexts and make simply retrieving a limited number of passages from contexts not sufficient for these tasks. Based on , we evaluate several state-of-the-art LLMs tailored for processing long contexts. The experimental results indicate that existing long-context LLMs still require significant advancements to process 100K+ contexts effectively. Furthermore, we present three intriguing analyses regarding the behavior of LLMs processing long context. Our code and data is released.

“跨语言代码生成旨在将英语到代码的生成能力迁移至其他自然语言。翻译-训 练(Translate-Train)和语码转换(Code-Switching)是实现跨语言迁移的两类经典数据增广方法,两者优势互补但尚未有效结合。为此,本文提出了一种面向跨语言代码生成的片段级语码转换(SpanCS)方法。首先,该方法利用语码转换框架关联源语言上下文与目标语言片段,以促进多种语言的交互和对齐。其次,该方法利用翻译-训练方法从完整的源语言翻译中提取目标语言片段,以保证增广数据与原始数据间的语义一致性。为了公平地评价多种自然语言之间代码生成的性能差异,本文通过人工翻译与校验,基于HumanEval构建了包含10种自然语言的多语言代码生成评测基MHumanEval。该基准上的三个主干模型的实验结果表明,SpanCS在跨语言代码生成任务上一致优于前人的数据增广方法。”

“尽管大语言模型在自然语言处理任务中取得显著进展,但其在复杂问题推理等领域还面临着认知负荷问题,即大语言模型在推理过程需要记忆并处理大量信息。因此,如何有效地减少语言模型推理过程中的认知负荷,缓解推理过程中可能出现的认知过载是一个亟待解决的问题。对此本文提出了Self-Guide方法,用于增强语言模型的推理能力。该方法通过指引大语言模型生成常识知识和推理指导,让语言模型基于自我规划来增强其推理能力,并通过与推理链结合的方式对模型的推理过程进行校准。与现有方法不同的是,本文在不对大语言模型进行微调或使用外部工具的情况下,显著提升了语言模型的推理性能。实验结果表明,Self-Guide方法在四种常见推理任务上性能显著优于基线方法,同时相比传统的推理链模型,Self-Guide方法在推理能力较弱的模型上也具有良好的泛化性能。通过结合大语言模型的自我规划和推理能力,Self-Guide方法为提升语言模型的推理能力提供了一种新的有效途径。”

“Free-form table question answering is a challenging task since tables contain structured contentscompared to plain texts, which requires high-level reasoning abilities to effectively identify cellsthat are relevant to the question and produce a correct and faithful answer based on their relations.Large language models (LLMs) have exhibited remarkable reasoning capabilities in numerousNLP applications. However, in some specific tasks, specially-trained small models can still out-perform LLMs. Furthermore, small models require extremely less computation costs comparedto LLMs. To leverage the strengths of both types of models, we propose a Relevant-Cell-basedKnowledge Distillation with inference-time Teacher Guidance (RCKD-TG) method. This ap-proach aims to combine small free-form table question answering models’ abilities to learn fromhuman annotations and large language models’ abilities to effectively reason from table contents,via applying Relevant-Cell-based rationales distilled from LLMs to small models’ training andinference stages. Our experiments demonstrate the superiority of our method over vanilla smallmodels in correctness, faithfulness, adequacy and fluency, also over general LLMs in adheringto the style of human annotations. We achieve state-of-the-art performance on FeTaQA, a rep-resentative free-form table question answering benchmark. Our result of a 41.3 BLEU scoredemonstrates the feasibility of effectively using small models’ task-specific abilities and LLMs’reasoning capabilities at the same time. Additionally, our method exhibits high computation ef-ficiency and data efficiency. Compared to strong baselines, we achieve better performance withsignificantly less training data.”

Text simplification aims to make the text easier to understand by applying rewriting transformations. There has been very little research on Chinese text simplification for a long time. The lack of generic evaluation data is an essential reason for this phenomenon. In this paper, we introduce MCTS, a multi-reference Chinese text simplification dataset. We describe the annotation process of the dataset and provide a detailed analysis. Furthermore, we evaluate the performance of several unsupervised methods and advanced large language models. We additionally provide Chinese text simplification parallel data that can be used for training, acquired by utilizing machine translation and English text simplification. We hope to build a basic understanding of Chinese text simplification through the foundational work and provide references for future research. All of the code and data are released at https://github.com/blcuicall/mcts/.

Although neural machine translation (NMT) models perform well in the general domain, it remains rather challenging to control their generation behavior to satisfy the requirement of different users. Given the expensive training cost and the data scarcity challenge of learning a new model from scratch for each user requirement, we propose a memory-augmented adapter to steer pretrained NMT models in a pluggable manner. Specifically, we construct a multi-granular memory based on the user-provided text samples and propose a new adapter architecture to combine the model representations and the retrieved results. We also propose a training strategy using memory dropout to reduce spurious dependencies between the NMT model and the memory. We validate our approach on both style- and domain-specific experiments and the results indicate that our method can outperform several representative pluggable baselines.

Explainable recommendation is a technique that combines prediction and generation tasks to produce more persuasive results. Among these tasks, textual generation demands large amounts of data to achieve satisfactory accuracy. However, historical user reviews of items are often insufficient, making it challenging to ensure the precision of generated explanation text. To address this issue, we propose a novel model, ERRA (Explainable Recommendation by personalized Review retrieval and Aspect learning). With retrieval enhancement, ERRA can obtain additional information from the training sets. With this additional information, we can generate more accurate and informative explanations. Furthermore, to better capture users’ preferences, we incorporate an aspect enhancement component into our model. By selecting the top-n aspects that users are most concerned about for different items, we can model user representation with more relevant details, making the explanation more persuasive. To verify the effectiveness of our model, extensive experiments on three datasets show that our model outperforms state-of-the-art baselines (for example, 3.4% improvement in prediction and 15.8% improvement in explanation for TripAdvisor).

As e-commerce platforms develop different business lines, a special but challenging product categorization scenario emerges, where there are multiple domain-specific category taxonomies and each of them evolves dynamically over time. In order to unify the categorization process and ensure efficiency, we propose a two-stage taxonomy-agnostic framework that relies solely on calculating the semantic relatedness between product titles and category names in the vector space. To further enhance domain transferability and better exploit cross-domain data, we design two plug-in modules: a heuristic mapping scorer and a pretrained contrastive ranking module with the help of meta concepts, which represent keyword knowledge shared across domains. Comprehensive offline experiments show that our method outperforms strong baselineson three dynamic multi-domain product categorization (DMPC) tasks,and online experiments reconfirm its efficacy with a5% increase on seasonal purchase revenue. Related datasets will be released.

Machine translation systems are expected to cope with various types of constraints in many practical scenarios. While neural machine translation (NMT) has achieved strong performance in unconstrained cases, it is non-trivial to impose pre-specified constraints into the translation process of NMT models. Although many approaches have been proposed to address this issue, most existing methods can not satisfy the following three desiderata at the same time: (1) high translation quality, (2) high match accuracy, and (3) low latency. In this work, we propose a template-based method that can yield results with high translation quality and match accuracy and the inference speed of our method is comparable with unconstrained NMT models. Our basic idea is to rearrange the generation of constrained and unconstrained tokens through a template. Our method does not require any changes in the model architecture and the decoding algorithm. Experimental results show that the proposed template-based approach can outperform several representative baselines in both lexically and structurally constrained translation tasks.

Confidence calibration, which aims to make model predictions equal to the true correctness measures, is important for neural machine translation (NMT) because it is able to offer useful indicators of translation errors in the generated output. While prior studies have shown that NMT models trained with label smoothing are well-calibrated on the ground-truth training data, we find that miscalibration still remains a severe challenge for NMT during inference due to the discrepancy between training and inference. By carefully designing experiments on three language pairs, our work provides in-depth analyses of the correlation between calibration and translation performance as well as linguistic properties of miscalibration and reports a number of interesting findings that might help humans better analyze, understand and improve NMT models. Based on these observations, we further propose a new graduated label smoothing method that can improve both inference calibration and translation performance.

While back-translation is simple and effective in exploiting abundant monolingual corpora to improve low-resource neural machine translation (NMT), the synthetic bilingual corpora generated by NMT models trained on limited authentic bilingual data are inevitably noisy. In this work, we propose to quantify the confidence of NMT model predictions based on model uncertainty. With word- and sentence-level confidence measures based on uncertainty, it is possible for back-translation to better cope with noise in synthetic bilingual corpora. Experiments on Chinese-English and English-German translation tasks show that uncertainty-based confidence estimation significantly improves the performance of back-translation.