For speaker recognition, pseudo-labeling has shown advantages in alleviating the scarcity of labeled data. Inspired by image classification tasks, existing methods typically adopt threshold-based strategies to identify reliable pseudo-labels. However, compared to image classification, speaker recognition requires finer-grained class discrimination for open-set identity verification, and thus often adopts margin-based losses that amplify gradients near decision boundaries. While effective under full supervision, this design increases sensitivity to noisy or sparse pseudo-labels, limiting the effectiveness of threshold-based selection. In this work, we propose SpeakerMatch , a novel distribution-based framework for semi-supervised and self-supervised speaker recognition. SpeakerMatch models the confidence distribution to distinguish reliable pseudo-labels from noisy ones globally and selects those whose confidence values and confidence prediction behaviors closely align with high-quality signals. Systematic evaluation across five settings shows that our method outperforms existing approaches, achieving a 13.7% relative improvement over the best semi-supervised speaker recognition baseline, while also delivering a lower equal error rate (EER) and reduced training costs compared to self-supervised methods.

Open-set speaker recognition is to identify whether the voices are from the same speaker. One challenge of speaker recognition is collecting large amounts of high-quality data. Based on the promising results of image classification, one intuitively feasible solution is semi-supervised learning (SSL) which uses confidence thresholds to assign pseudo labels for unlabeled data. However, we empirically demonstrated that applying SSL methods to speaker recognition is non-trivial. These methods focus solely on inter-class discrepancy as thresholds to select pseudo labels, overlooking intra-class compactness, which is particularly important for open-set speaker recognition tasks. Motivated by this, we propose Int*-Match, a semi-supervised speaker recognition method selecting reliable pseudo labels with intra-class compactness and inter-class discrepancy for speaker recognition. In particular, we use the inter-class discrepancy of labeled data as the threshold for pseudo-label selection and adjust the threshold based on the intra-class compactness of the pseudo labels dynamically and adaptively. Our systematic experiments demonstrate the superiority of Int*-Match, presenting an outstanding Equal Error Rate (EER) of 1.00% on the VoxCeleb1 original test set, which is merely 0.06% below the performance achieved by fully supervised learning.

Learning discriminative representations of different speakers is a key challenge in open-set speaker recognition. To mitigate the mismatch between closed-set training and open-set testing, margin-based losses have been widely adopted to directly optimize the cosine similarity between speaker representations and proxy class vectors. While recent studies have shown that enhancing the margin for hard samples can improve representation learning, we observe three key limitations: (1) the measurement of sample hardness fails to fully capture differences in speaker representations, (2) margin-based emphasis does not significantly increase the gradient magnitude, and (3) the potential performance degradation caused by emphasizing hard samples are rarely considered. To address these issues, we propose Adaspeaker, a novel loss framework that combines an Intra-Inter sample hardness coefficient (Int2H) with a gradient-aware adaptive scaling strategy. Specifically, Int2H jointly models inter-class and intra-class hardness to estimate sample importance, which is subsequently used to adaptively scale cosine similarities for enhancing the gradient contribution of important samples. Experiments conducted on five evaluation settings show that Adaspeaker outperforms existing loss functions. Moreover, Adaspeaker can be seamlessly integrated into margin-based losses, yielding an average performance improvement of 12.6%. Code is available at https://github.com/LiuJinghan2001/Adaspeaker.

Holdout validation and hyperparameter tuning from data is a long-standing problem in offline reinforcement learning (RL). A standard framework is to use off-policy evaluation (OPE) methods to evaluate and select the policies, but OPE either incurs exponential variance (e.g., importance sampling) or has hyperparameters on their own (e.g., FQE and model-based). We focus on hyperparameter tuning for OPE itself, which is even more under-investigated. Concretely, we select among candidate value functions ("model-free") or dynamics models ("model-based") to best assess the performance of a target policy. We develop: (1) new model-free and model-based selectors with theoretical guarantees, and (2) a new experimental protocol for empirically evaluating them. Compared to the model-free protocol in prior works, our new protocol allows for more stable generation and better control of candidate value functions in an optimization-free manner, and evaluation of model-free and model-based methods alike. We exemplify the protocol on Gym-Hopper, and find that our new model-free selector, LSTD-Tournament, demonstrates promising empirical performance.

Digital advertising constitutes one of the main revenue sources for online platforms. In recent years, some advertisers tend to adopt auto-bidding tools to facilitate advertising performance optimization, making the classical utility maximizer model in auction theory not fit well. Some recent studies proposed a new model, called value maximizer, for auto-bidding advertisers with return-on-investment (ROI) constraints. However, the model of either utility maximizer or value maximizer could only characterize partial advertisers in real-world advertising platforms. In a mixed environment where utility maximizers and value maximizers coexist, the truthful ad auction design would be challenging since bidders could manipulate both their values and affiliated classes, leading to a multi-parameter mechanism design problem. In this work, we address this issue by proposing a payment rule which combines the corresponding ones in classical VCG and GSP mechanisms in a novel way. Based on this payment rule, we propose a truthful auction mechanism with an approximation ratio of 2 on social welfare, which is close to the lower bound of at least 5/4 that we also prove. The designed auction mechanism is a generalization of VCG for utility maximizers and GSP for value maximizers.

Information literacy is a growing requirement for all majors, while engineering students have a stronger interest in developing this skill. Lead Users, who are proficient in information and digital literacy, are particularly adept at their profession. To address this, our university's subject librarian explored the expertise of Lead Users and developed an “Information Talent” program to facilitate interdisciplinary academic sharing. The program covers a wide range of topics and provides students with advanced information literacy skills and knowledge that cannot be acquired through traditional lectures. By using Lead Users as resources and supplementing original information literacy education, the program aims to improve students' information literacy. We have implemented a series of flexible, interactive, innovative activities that have enriched our university's information literacy education system and enhanced subject services on the new academic platform. These activities have been integrated into the university's academic ecology and received widespread recognition for supporting student personnel training. To ensure the sustainability of the program, we have incorporated the database system of “Information Talent”, which can sort out activity details by topic, talent, type, scope, and other keywords. Through analysis of the database system, we can effectively grasp the key details of activities and accumulate sufficient resources to guide future activities, ensuring the sustainable development of the program. This paper analyzes the practice of the “Information Talent” sharing program in detail, including key points and difficulties in the process, as well as solutions and ideas. Additionally, it proposes some ideas for future activities and sustainable development.

This Research-to-Practice Work in Progress paper presents DCLab, a web-based system for conducting digital logic experiments online, to improve both the effectiveness and the efficiency of digital logic experiment teaching. DCLab covers all experimental contents required by traditional digital logic experiment classes. It allows the students to draw circuit diagrams or to write VHDL code to design their own circuits, and it provides complete simulation functions. DCLab records the progress of each student and makes it convenient for a student to review the history of his practice. Furthermore, for the instructors, they are able to post homework assignments on DCLab, and the system will automatically add homework projects to the students' home pages. Statistics about how the students perform on the homework will be displayed to the instructors, which may help them develop more effective courses. We have tested DCLab among students in the digital logic course, and the results have confirmed its validity.