Yan Zhang

Hao Sun, Xiao Liu, Yeyun Gong et al. · microsoft-research, pku

Knowledge distillation is often used to transfer knowledge from a strong teacher model to a relatively weak student model. Traditional methods include response-based methods and feature-based methods. Response-based methods are widely used but suffer from lower upper limits of performance due to their ignorance of intermediate signals, while feature-based methods have constraints on vocabularies, tokenizers and model architectures. In this paper, we propose a liberal feature-based distillation method (LEAD). LEAD aligns the distribution between the intermediate layers of teacher model and student model, which is effective, extendable, portable and has no requirements on vocabularies, tokenizers, or model architectures. Extensive experiments show the effectiveness of LEAD on widely-used benchmarks, including MS MARCO Passage Ranking, TREC 2019 DL Track, MS MARCO Document Ranking and TREC 2020 DL Track. Our code is available in https://github.com/microsoft/SimXNS/tree/main/LEAD.

Haoran Ma, Yan Zhang, Pengyuan Liu et al.

The Attention Restoration Theory (ART) presents a theoretical framework with four essential indicators (being away, extent, fascinating, and compatibility) for comprehending urban and natural restoration quality. However, previous studies relied on non-sequential data and non-spatial dependent methods, which overlooks the impact of spatial structure defined here as the positional relationships between scene entities on restoration quality. The past methods also make it challenging to measure restoration quality on an urban scale. In this work, a spatial-dependent graph neural networks (GNNs) approach is proposed to reveal the relation between spatial structure and restoration quality on an urban scale. Specifically, we constructed two different types of graphs at the street and city levels. The street-level graphs, using sequential street view images (SVIs) of road segments to capture position relationships between entities, were used to represent spatial structure. The city-level graph, modeling the topological relationships of roads as non-Euclidean data structures and embedding urban features (including Perception-features, Spatial-features, and Socioeconomic-features), was used to measure restoration quality. The results demonstrate that: 1) spatial-dependent GNNs model outperforms traditional methods (Acc = 0.735, F1 = 0.732); 2) spatial structure portrayed through sequential SVIs data significantly influences restoration quality; 3) spaces with the same restoration quality exhibited distinct spatial structures patterns. This study clarifies the association between spatial structure and restoration quality, providing a new perspective to improve urban well-being in the future.

Timin Gao, Peixian Chen, Mengdan Zhang et al.

With the advent of large language models(LLMs) enhanced by the chain-of-thought(CoT) methodology, visual reasoning problem is usually decomposed into manageable sub-tasks and tackled sequentially with various external tools. However, such a paradigm faces the challenge of the potential "determining hallucinations" in decision-making due to insufficient visual information and the limitation of low-level perception tools that fail to provide abstract summaries necessary for comprehensive reasoning. We argue that converging visual context acquisition and logical reasoning is pivotal for tackling visual reasoning tasks. This paper delves into the realm of multimodal CoT to solve intricate visual reasoning tasks with multimodal large language models(MLLMs) and their cognitive capability. To this end, we propose an innovative multimodal CoT framework, termed Cantor, characterized by a perception-decision architecture. Cantor first acts as a decision generator and integrates visual inputs to analyze the image and problem, ensuring a closer alignment with the actual context. Furthermore, Cantor leverages the advanced cognitive functions of MLLMs to perform as multifaceted experts for deriving higher-level information, enhancing the CoT generation process. Our extensive experiments demonstrate the efficacy of the proposed framework, showing significant improvements in multimodal CoT performance across two complex visual reasoning datasets, without necessitating fine-tuning or ground-truth rationales. Project Page: https://ggg0919.github.io/cantor/ .

Danqing Luo, Chen Zhang, Yan Zhang et al.

Training or finetuning large-scale language models (LLMs) requires substantial computation resources, motivating recent efforts to explore parameter-efficient adaptation to downstream tasks. One approach is to treat these models as black boxes and use forward passes (Inference APIs) to interact with them. Current research focuses on adapting these black-box models to downstream tasks using gradient-free prompt optimization, but this often involves an expensive process of searching task-specific prompts. Therefore, we are motivated to study black-box language model adaptation without prompt search. Specifically, we introduce a label-enhanced cross-attention network called CrossTune, which models the semantic relatedness between the input text sequence and task-specific label descriptions. Its effectiveness is examined in the context of few-shot text classification. To improve the generalization of CrossTune, we utilize ChatGPT to generate additional training data through in-context learning. A switch mechanism is implemented to exclude low-quality ChatGPT-generated data. Through extensive experiments on seven benchmark text classification datasets, we demonstrate that our proposed approach outperforms the previous state-of-the-art gradient-free black-box tuning method by 5.7% on average. Even without using ChatGPT-augmented data, CrossTune performs better or comparably than previous black-box tuning methods, suggesting the effectiveness of our approach.

Yu Zhang, Wei Wei, Binxuan Huang et al.

Real-time location inference of social media users is the fundamental of some spatial applications such as localized search and event detection. While tweet text is the most commonly used feature in location estimation, most of the prior works suffer from either the noise or the sparsity of textual features. In this paper, we aim to tackle these two problems. We use topic modeling as a building block to characterize the geographic topic variation and lexical variation so that "one-hot" encoding vectors will no longer be directly used. We also incorporate other features which can be extracted through the Twitter streaming API to overcome the noise problem. Experimental results show that our RATE algorithm outperforms several benchmark methods, both in the precision of region classification and the mean distance error of latitude and longitude regression.

Heng Zhang, Yan Zhang, Jia-Huai You et al.

An ontology language for ontology mediated query answering (OMQA-language) is universal for a family of OMQA-languages if it is the most expressive one among this family. In this paper, we focus on three families of tractable OMQA-languages, including first-order rewritable languages and languages whose data complexity of the query answering is in AC0 or PTIME. On the negative side, we prove that there is, in general, no universal language for each of these families of languages. On the positive side, we propose a novel property, the locality, to approximate the first-order rewritability, and show that there exists a language of disjunctive embedded dependencies that is universal for the family of OMQA-languages with locality. All of these results apply to OMQA with query languages such as conjunctive queries, unions of conjunctive queries and acyclic conjunctive queries.

Vernon Asuncion, Yan Zhang

In this paper we introduce a new class of tuple-generating dependencies (TGDs) called triangularly-guarded (TG) TGDs. We show that conjunctive query answering under this new class of TGDs is decidable since this new class of TGDs also satisfies the finite controllability (FC) property. We further show that this new class strictly contains some other decidable classes such as weak-acyclic, guarded, sticky and shy. In this sense, the class TG provides a unified representation of all these aforementioned classes of TGDs.

Vernon Asuncion, Yan Zhang, Heng Zhang et al.

In ontology-based data access (OBDA), the classical database is enhanced with an ontology in the form of logical assertions generating new intensional knowledge. A powerful form of such logical assertions is the tuple-generating dependencies (TGDs), also called existential rules, where Horn rules are extended by allowing existential quantifiers to appear in the rule heads. In this paper we introduce a new language called loop restricted (LR) TGDs (existential rules), which are TGDs with certain restrictions on the loops embedded in the underlying rule set. We study the complexity of this new language. We show that the conjunctive query answering (CQA) under the LR TGDs is decid- able. In particular, we prove that this language satisfies the so-called bounded derivation-depth prop- erty (BDDP), which implies that the CQA is first-order rewritable, and its data complexity is in AC0 . We also prove that the combined complexity of the CQA is EXPTIME complete, while the language membership is PSPACE complete. Then we extend the LR TGDs language to the generalised loop restricted (GLR) TGDs language, and prove that this class of TGDs still remains to be first-order rewritable and properly contains most of other first-order rewritable TGDs classes discovered in the literature so far.

Vernon Asuncion, Yan Zhang

In this paper we introduce a new class of tuple-generating dependencies (TGDs) called triangularly-guarded TGDs, which are TGDs with certain restrictions on the atomic derivation track embedded in the underlying rule set. We show that conjunctive query answering under this new class of TGDs is decidable. We further show that this new class strictly contains some other decidable classes such as weak-acyclic, guarded, sticky and shy, which, to the best of our knowledge, provides a unified representation of all these aforementioned classes.

Xing Zhang, Zhenglei Yi, Zhi Yan et al.

Mobile big data contains vast statistical features in various dimensions, including spatial, temporal, and the underlying social domain. Understanding and exploiting the features of mobile data from a social network perspective will be extremely beneficial to wireless networks, from planning, operation, and maintenance to optimization and marketing. In this paper, we categorize and analyze the big data collected from real wireless cellular networks. Then, we study the social characteristics of mobile big data and highlight several research directions for mobile big data in the social computing areas.

Heng Zhang, Yan Zhang, Jia-Huai You

Existential rules, also known as data dependencies in Databases, have been recently rediscovered as a promising family of languages for Ontology-based Query Answering. In this paper, we prove that disjunctive embedded dependencies exactly capture the class of recursively enumerable ontologies in Ontology-based Conjunctive Query Answering (OCQA). Our expressive completeness result does not rely on any built-in linear order on the database. To establish the expressive completeness, we introduce a novel semantic definition for OCQA ontologies. We also show that neither the class of disjunctive tuple-generating dependencies nor the class of embedded dependencies is expressively complete for recursively enumerable OCQA ontologies.

Hai Wan, Heng Zhang, Peng Xiao et al.

Traditional inconsistency-tolerent query answering in ontology-based data access relies on selecting maximal components of an ABox/database which are consistent with the ontology. However, some rules in ontologies might be unreliable if they are extracted from ontology learning or written by unskillful knowledge engineers. In this paper we present a framework of handling inconsistent existential rules under stable model semantics, which is defined by a notion called rule repairs to select maximal components of the existential rules. Surprisingly, for R-acyclic existential rules with R-stratified or guarded existential rules with stratified negations, both the data complexity and combined complexity of query answering under the rule {repair semantics} remain the same as that under the conventional query answering semantics. This leads us to propose several approaches to handle the rule {repair semantics} by calling answer set programming solvers. An experimental evaluation shows that these approaches have good scalability of query answering under rule repairs on realistic cases.