Model calibration aligns a model’s confidence with the accuracy of its predictions, directly influencing reliability and user trust. Surprisingly, existing research on knowledge distillation has paid little attention to the calibration of student models. In this paper, we evaluate calibration across diverse knowledge distillation settings and demonstrate that student models distilled from black-box teachers generally exhibit poor calibration. We attribute this to the fact that distillation fails to convey well-calibrated confidence, which serves as a crucial expression of predictive uncertainty. To address this limitation, we propose an ***U**ncertainty-**I**ntegrated **K**nowledge **D**istillation* (**UIKD**) method comprising two key components: (1) *uncertainty estimation*, which leverages the collective intelligence of multiple LLMs via a debate mechanism to quantify uncertainty in the distillation data; and (2) *uncertainty integration*, which employs an adaptive weighted loss function to integrate this data uncertainty into the student model effectively. Experiments on common-sense reasoning and natural language inference benchmarks demonstrate that UIKD consistently improves student model calibration while maintaining predictive accuracy. Our code is available at https://github.com/RuixuanZhang127/UIKD.

Document ranking aims to order candidate documents based on their relevance to a given query. However, identifying relevant documents in complex real-world scenarios requires deep reasoning beyond surface-level matching, driving a shift from semantic matching to reasoning-intensive reranking. Since the latter requires integrating information across documents for complex reasoning, it presents two key challenges: (1) varying reasoning difficulty, where assessing document relevance to the query requires multi-step reasoning with different complexity; and (2) strong contextual sensitivity, where the relevance of a document is heavily influenced by other documents within the context window. To address these challenges, we propose CuRLRank, which consists of (1) CuRL, an adaptive curriculum reinforcement learning approach that enables LLMs to develop complex reasoning capabilities across varying levels of difficulty, and (2) IFRS, an iterative filtering ranking strategy to achieve efficient and robust ranking. Extensive experiments on BRIGHT and R2MED benchmarks demonstrate that our CuRLRank significantly outperforms baseline methods while achieving lower ranking latency.

The task of model editing becomes popular for correcting inaccurate or outdated parametric knowledge in Large Language Models (LLMs). However, there are major limitations of state of the art (SOTA) model editing methods, including the excessive memorization issue caused by the direct editing methods, as well as the error propagation and knowledge conflict issues from the memory enhancement methods, resulting in hindering models’ portability, e.g., the ability to transfer the new knowledge to related one-hop or multi-hop content. To address these issues, we propose the InstructEd method, the idea of which is to insert soft instructions into the attention module so as to facilitate interactions between instructions and questions and to understand and utilize new facts. Our main findings are: (i) InstructEd has achieved SOTA performance on three datasets for one-hop/multi-hop evaluation with LLaMAs and GPT2, achieving 10% (5%) improvement in one-hop (multi-hop) model editing.(ii) Different from earlier methods on editing parameters in FFN, we show that editing attention can also help. (iii) Model editing is highly related to retrieval augmented methods, which can help improve the locality of model editing while slightly decrease the editing performance with hops.

The task of sequential model editing is to fix erroneous knowledge in Pre-trained Language Models (PLMs) efficiently, precisely and continuously. Although existing methods can deal with a small number of modifications, these methods experience a performance decline or require additional annotated data, when the number of edits increases. In this paper, we propose a Retrieval Augmented Sequential Model Editing framework (RASE) that leverages factual information to enhance editing generalization and to guide the identification of edits by retrieving related facts from the fact-patch memory we constructed. Our main findings are: (i) State-of-the-art models can hardly correct massive mistakes stably and efficiently; (ii) Even if we scale up to thousands of edits, RASE can significantly enhance editing generalization and maintain consistent performance and efficiency; (iii) RASE can edit large-scale PLMs and increase the performance of different editors. Moreover, it can integrate with ChatGPT and further improve performance. Our code and data are available at: https://github.com/sev777/RASE.

Frame Identification (FI) is a fundamental and challenging task in frame semantic parsing. The task aims to find the exact frame evoked by a target word in a given sentence. It is generally regarded as a classification task in existing work, where frames are treated as discrete labels or represented using onehot embeddings. However, the valuable knowledge about frames is neglected. In this paper, we propose a Knowledge-Guided Frame Identification framework (KGFI) that integrates three types frame knowledge, including frame definitions, frame elements and frame-to-frame relations, to learn better frame representation, which guides the KGFI to jointly map target words and frames into the same embedding space and subsequently identify the best frame by calculating the dot-product similarity scores between the target word embedding and all of the frame embeddings. The extensive experimental results demonstrate KGFI significantly outperforms the state-of-the-art methods on two benchmark datasets.

框架语义角色标注(Frame Semantic Role Labeling, FSRL)是基于FrameNet标注体系的语义分析任务。语义角色标注通常对句法有很强的依赖性,目前的语义角色标注模型大多基于双向长短时记忆网络Bi-LSTM,虽然可以获取句子中的长距离依赖信息,但无法很好获取句子中的句法信息。因此,引入self-attention机制来捕获句子中每个词的句法信息。实验结果表明,该模型在CFN(Chinese FrameNet,汉语框架网)数据集上的F1达到83.77%,提升了近11%。