Zhang Min

• PhD
• Professor at Xidian University

Legal case retrieval aims to help legal workers find relevant cases related to their cases at hand, which is important for the guarantee of fairness and justice in legal judgments. While recent advances in neural retrieval methods have significantly improved the performance of open-domain retrieval tasks (e.g., Web search), their advantages haven’t...

The emergence of agentic recommender systems powered by Large Language Models (LLMs) represents a paradigm shift in personalized recommendations, leveraging LLMs' advanced reasoning and role-playing capabilities to enable autonomous, adaptive decision-making. Unlike traditional recommendation approaches, agentic recommender systems can dynamically...

The rapid growth of short videos has necessitated effective recommender systems to match users with content tailored to their evolving preferences. Current video recommendation models primarily treat each video as a whole, overlooking the dynamic nature of user preferences with specific video segments. In contrast, our research focuses on segment-l...

Language reconstruction from non-invasive brain recordings has been a long-standing challenge. Existing research has addressed this challenge with a classification setup, where a set of language candidates are pre-constructed and then matched with the representation decoded from brain recordings. Here, we propose a method that addresses language re...

Intelligence is a crucial trait for species to find solutions within a limited number of trial-and-error attempts. Building on this idea, we introduce Survival Game as a framework to evaluate intelligence based on the number of failed attempts in a trial-and-error process. Fewer failures indicate higher intelligence. When the expectation and varian...

With the rapid development of Large Language Models (LLMs), recent studies employed LLMs as recommenders to provide personalized information services for distinct users. Despite efforts to improve the accuracy of LLM-based recommendation models, relatively little attention is paid to beyond-utility dimensions. Moreover, there are unique evaluation...

With the rapid development of large language models (LLMs), how to efficiently evaluate them has become an important research question. Existing evaluation methods often suffer from high costs, limited test formats, the need of human references, and systematic evaluation biases. To address these limitations, our study introduces the Auto-PRE, an au...

Despite having powerful reasoning and inference capabilities, Large Language Models (LLMs) still need external tools to acquire real-time information retrieval or domain-specific expertise to solve complex tasks, which is referred to as tool learning. Existing tool learning methods primarily rely on tuning with expert trajectories, focusing on toke...

Legal case retrieval aims to help legal workers find relevant cases related to their cases at hand, which is important for the guarantee of fairness and justice in legal judgments. While recent advances in neural retrieval methods have significantly improved the performance of open-domain retrieval tasks (e.g., Web search), their advantages haven't...

Providing reasonable explanations for a specific suggestion given by the recommender can help users trust the system more. As logic rule-based inference is concise, transparent, and aligned with human cognition, it can be adopted to improve the interpretability of recommendation models. Previous work that interprets user preference with logic rules...

Recent advances in Large Language Models (LLMs) have significantly shaped the applications of AI in multiple fields, including the studies of legal intelligence. Trained on extensive legal texts, including statutes and legal documents, the legal LLMs can capture important legal knowledge/concepts effectively and provide important support for downst...

Language reconstruction from non-invasive brain recordings has been a long-standing challenge. Existing research has addressed this challenge with a classification setup, where a set of language candidates are pre-constructed and then matched with the representation decoded from brain recordings. Here, we propose a new method that addresses languag...

With the applications of recommendation systems rapidly expanding, an increasing number of studies have focused on every aspect of recommender systems with different data inputs, models, and task settings. Therefore, a flexible library is needed to help researchers implement the experimental strategies they require. Existing open libraries for reco...

The Relevance Feedback (RF) process relies on accurate and real-time relevance estimation of feedback documents to improve retrieval performance. Since collecting explicit relevance annotations imposes an extra burden on the user, extensive studies have explored using pseudo-relevance signals and implicit feedback signals as substitutes. However, s...

Dense Retrieval~(DR) has achieved state-of-the-art first-stage ranking effectiveness. However, the efficiency of most existing DR models is limited by the large memory cost of storing dense vectors and the time-consuming nearest neighbor search~(NNS) in vector space. Therefore, we present RepCONC, a novel retrieval model that learns discrete Repres...

Legal case retrieval has received increasing attention in recent years. However, compared to ad-hoc retrieval tasks, legal case retrieval has its unique challenges. First, case documents are rather lengthy and contain complex legal structures. Therefore, it is difficult for most existing dense retrieval models to encode an entire document and captu...

In the era of information explosion, numerous items emerge every day, especially in feed scenarios. Due to the limited system display slots and user browsing attention, various recommendation systems are designed not only to satisfy users' personalized information needs but also to allocate items' exposure. However, recent recommendation studies ma...

Our team(THUIR2) participated in both FOSS and POSS subtasks of the NTCIR-161 Session Search (SS) Task. This paper describes our approaches and results. In the FOSS subtask, we submit five runs using learning-to-rank and fine-tuned pre-trained language models. We fine-tuned the pre-trained language model with ad-hoc data and session information and...

Background As rare diseases (RDs) receive increasing attention, obtaining accurate RD incidence estimates has become an essential concern in public health. Since RDs are difficult to diagnose, include diverse types, and have scarce cases, traditional epidemiological methods are costly in RD registries. With the development of the internet, users ha...

Ranking ensemble is a critical component in real recommender systems. When a user visits a platform, the system will prepare several item lists, each of which is generally from a single behavior objective recommendation model. As multiple behavior intents, e.g., both clicking and buying some specific item category, are commonly concurrent in a user...

Legal case retrieval is a specialized IR task aiming to retrieve supporting cases given a query case. Existing work has shown that the conversational search paradigm can improve users’ search experience in legal case retrieval with humans as intermediary agents. To move further towards a practical system, it is essential to decide what action a com...

Recently, pre-trained language models (PTM) have achieved great success on ad hoc search. However, the performance decline in low-resource scenarios demonstrates the capability of PTM has not been inspired fully. As a novel paradigm to apply PTM to downstream tasks, prompt learning is a feasible scheme to boost PTM’s performance by aligning the pre...

The trimaran is an advanced ship with high hydrodynamic performance, but the study of its electromagnetic characteristics and Doppler spectrum is insufficient. This study presents a new procedure for evaluating the scattering echo and Doppler spectrum of a trimaran in a time-varying ocean environment. First, based on the Reynolds-averaged Navier-St...

As recommender systems become increasingly important in daily human decision-making, users are demanding convincing explanations to understand they get the specific recommendation results. Although a number of explainable recommender systems have recently been proposed, there still lacks an understanding of what users really need in a recommendatio...

Traditional wake detection methods have been successfully applied to the detection of a simple linear ship wake. However, they cannot effectively detect nonlinear wake and weak wake under high sea state conditions, whereas the deep-learning-based detection method could play to its strengths in this respect. Due to the lack of sufficient measured SA...

Dense retrieval models represent queries and documents with one or multiple fixed-width vectors and retrieve relevant documents via nearest neighbor search. Recently these models have shown improvement in retrieval performance and have drawn increasing attention from the IR community. Among a variety of dense retrieval models, the models that emplo...

BACKGROUND As rare diseases (RDs) receive increasing attention, obtaining accurate RD incidence estimates has become an essential concern in public health. Since RDs are difficult to diagnose, include diverse types, and have scarce cases, traditional epidemiological methods are costly in RD registries. With the development of the internet, users ha...

Nowadays, recommender systems play an increasingly important role in the music scenario. Generally, music preferences are related to internal and external conditions. For example, mood state and ongoing activity will affect users' music preferences. However, conventional music recommenders cannot capture these conditions since they only utilize the...

In legal case retrieval, existing work has shown that human-mediated conversational search can improve users’ search experience. In practice, a suitable workflow can provide guidelines for constructing a machine-mediated agent replacing of human agents. Therefore, we conduct a comparison analysis and summarize two challenges when directly applying...

With the growth of information on the Web, most users heavily rely on information access systems (e.g., search engines, recommender systems, etc.) in their daily lives. During this procedure, modeling users' satisfaction status plays an essential part in improving their experiences with the systems. In this paper, we aim to explore the benefits of...

Recent advance in Dense Retrieval (DR) techniques has significantly improved the effectiveness of first-stage retrieval. Trained with large-scale supervised data, DR models can encode queries and documents into a low-dimensional dense space and conduct effective semantic matching. However, previous studies have shown that the effectiveness of DR mo...

Sentiment analysis is an essential task in natural language processing researches. Although existing works have gained much success with both statistical and neural-based solutions, little is known about the human decision process while performing this kind of complex cognitive task. Considering recent advances in human-inspired model design for NL...

Recommender systems are an essential tool to relieve the information overload challenge and play an important role in people’s daily lives. Since recommendations involve allocations of social resources (e.g., job recommendation), an important issue is whether recommendations are fair. Unfair recommendations are not only unethical but also harm the...

Dense Retrieval (DR) has achieved state-of-the-art first-stage ranking effectiveness. However, the efficiency of most existing DR models is limited by the large memory cost of storing dense vectors and the time-consuming nearest neighbor search (NNS) in vector space. Therefore, we present RepCONC, a novel retrieval model that learns discrete Repres...

Collaborative filtering (CF) plays a critical role in the development of recommender systems. Most CF methods utilize an encoder to embed users and items into the same representation space, and the Bayesian personalized ranking (BPR) loss is usually adopted as the objective function to learn informative encoders. Existing studies mainly focus on de...

The electromagnetic (EM) scattering from nearshore water waves impacted by gradually-varied underwater topography at UHF-band is numerically investigated using the second-order small slope approximation (SSA-II). Based on the fully nonlinear Boussinesq equation solver, the sea wave-like sinusoidal-type waves are generated and simulated in the shoal...

Recommender systems are an essential tool to relieve the information overload challenge and play an important role in people's daily lives. Since recommendations involve allocations of social resources (e.g., job recommendation), an important issue is whether recommendations are fair. Unfair recommendations are not only unethical but also harm the...

The micro-Doppler (m-D) signature can describe the motion state and structure of the target in detail. It can be considered as an effective method for ship target detection and recognition in the marine environment. In this paper, an effective strategy is proposed to simulate the m-D signature and extract the ship’s motion parameters on the time-va...

Chaff jamming is a widely used passive interference method. A chaff cloud diffusion model for the widespread chaff cloud was presented. The high-density chaff cloud aerodynamic model can rapidly predict the chaff elements’ time-varying spatial orientation, location, and overall spatial distribution. Basing on these pieces of information, the radar...

Recommendation in legal scenario (Legal-Rec) is a specialized recommendation task that aims to provide potential helpful legal documents for users. While there are mainly three differences compared with traditional recommendation: (1) Both the structural connections and textual contents of legal information are important in the Legal-Rec scenario,...

A retrieval model should not only interpolate the training data but also extrapolate well to the queries that are rather different from the training data. While dense retrieval (DR) models have been demonstrated to achieve better retrieval performance than the traditional term-based retrieval models, we still know little about whether they can extr...

Conversational Search has been paid much attention recently with the increasing popularity of intelligent user interfaces. However, compared with the endeavour in designing effective conversational search algorithms, relatively much fewer researchers have focused on the construction of benchmark datasets. For most existing datasets, the information...

Overfitting is a common problem in machine learning, which means the model too closely fits the training data while performing poorly in the test data. Among various methods of coping with overfitting, dropout is one of the representative ways. From randomly dropping neurons to dropping neural structures, dropout has achieved great success in impro...

Recommendation systems play a vital role in alleviating information overload. Generally, a recommendation model is trained to discern between positive (liked) and negative (disliked) instances for each user. However, under the open-world assumption, there are only positive instances but no negative instances from users’ implicit feedback, which pos...

Modern search engine result pages (SERPs) become increasingly complex with heterogeneous information aggregated from various sources. In many cases, these SERPs also display results in the right rail besides the traditional left-rail result lists, which change the linear result list to a non-linear panel and might influence user search behavior pat...

The rice extrudates obtained by varying feed moisture were used to evaluate the relationship between the mashing efficiency and physicochemical properties of adjuncts, including functional, rheological, pasting, and structural features. The RVA, XRD, FTIR, and rheological properties of broken rice flour and rice extrudates showed that extrusion wea...

Ultrasound at an intensity of 17.5, 20.0, 22.5, 25.0 and 27.5 W/L was used to assist dough fermentation to prepare steamed bread with 50% sweet potato pulp (SB-50% SPP), which was compared with SB-50% SPP without ultrasonic treatment. The dough rheology, starch–gluten network, texture characteristics and sensory quality of steamed bread with differ...

Differences in the motion of different parts of a target cause the echo signal to contain specific Doppler modulation information, i.e., the micro-Doppler (m-D) effect. This phenomenon provides an effective way to detect targets in marine environments. In this study, based on the establishment of the micromotion model of a rotating surveillance rad...

Overfitting is a common problem in machine learning, which means the model too closely fits the training data while performing poorly in the test data. Among various methods of coping with overfitting, dropout is one of the representative ways. From randomly dropping neurons to dropping neural structures, dropout has achieved great success in impro...

The electromagnetic scattering study of the turbulent wake of a moving ship has important application value in target recognition and tracking. However, to date, there has been insufficient research into the electromagnetic characteristics of near-field propeller turbulence. This study presents a new procedure for evaluating the electromagnetic sca...

With the realization of global navigation satellite system (GNSS) completion, GNSS reflectometry (GNSS-R) has become increasingly popular due to the advantages of global coverage and the availability of multiple sources in terms of earth remote sensing. This paper analyzes the Beidou navigation satellite system (BDS) signal reflection detection of...

Previous studies have demonstrated the potential bias and fairness issues in real recommender systems. Fairness issue is generally defined as equality of services between groups. However, fairness does not imply equality for all users/items in real scenarios, as premium users/items, who have paid for their services, are supposed to have better expe...

Dense Retrieval (DR) reaches state-of-the-art results in first-stage retrieval, but little is known about the mechanisms that contribute to its success. Therefore, in this work, we conduct an interpretation study of recently proposed DR models. Specifically, we first discretize the embeddings output by the document and query encoders. Based on the...

Apple pomace (AP) is often used directly as animal feed, while the value of feeding is limited by its low protein content. In this study, extrusion pretreatment was performed for AP, and further fermentation was carried out to improve its nutrition value. Strains suitable for extruded apple pomace (EAP) to produce high-quality microbial protein (MP...

A comprehensive electromagnetic scattering model for ship wakes on the sea surface is proposed to study the synthetic aperture radar (SAR) imagery for ship wakes. Our model considers a coupling of various wave systems, including Kelvin wake, turbulent wake, and the ocean ambient waves induced by the local wind. The fluid–structure coupling between...

While batch evaluation plays a central part in Information Retrieval (IR) research, most evaluation metrics are based on user models which mainly focus on browsing and clicking behaviors. As users' perceived satisfaction may also be impacted by their search intent, constructing different user models across various search intent may help design bett...