Author name cluster

Ju Fan

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

2 author rows

ICML Conference 2025 Conference Paper

Alpha-SQL: Zero-Shot Text-to-SQL using Monte Carlo Tree Search

Boyan Li 0001
Jiayi Zhang 0017
Ju Fan
Yanwei Xu 0004
Chong Chen 0001
Nan Tang 0001
Yuyu Luo

Text-to-SQL, which enables natural language interaction with databases, serves as a pivotal method across diverse industries. With new, more powerful large language models (LLMs) emerging every few months, fine-tuning has become incredibly costly, labor-intensive, and error-prone. As an alternative, zero-shot Text-to-SQL, which leverages the growing knowledge and reasoning capabilities encoded in LLMs without task-specific fine-tuning, presents a promising and more challenging direction. To address this challenge, we propose Alpha-SQL, a novel approach that leverages a Monte Carlo Tree Search (MCTS) framework to iteratively infer SQL construction actions based on partial reasoning states. To enhance the framework’s reasoning capabilities, we introduce LLM-as-Action-Model to dynamically generate SQL construction actions during the MCTS process, steering the search toward more promising SQL queries. Moreover, Alpha-SQL employs a self-supervised reward function to evaluate the quality of candidate SQL queries, ensuring more accurate and efficient query generation. Experimental results show that Alpha-SQL achieves 69. 7% execution accuracy on the BIRD development set, using a 32B open-source LLM without fine-tuning. Alpha-SQL outperforms the best previous zero-shot approach based on GPT-4o by 2. 5% on the BIRD development set.

Details

ICLR Conference 2025 Conference Paper

Harnessing Diversity for Important Data Selection in Pretraining Large Language Models

Chi Zhang 0102
Huaping Zhong
Kuan Zhang
Chengliang Chai
Rui Wang 0119
Xinlin Zhuang
Tianyi Bai
Jiantao Qiu

Data selection is of great significance in pretraining large language models, given the variation in quality within the large-scale available training corpora. To achieve this, researchers are currently investigating the use of data influence to measure the importance of data instances, $i.e.,$ a high influence score indicates that incorporating this instance to the training set is likely to enhance the model performance. Consequently, they select the top-$k$ instances with the highest scores. However, this approach has several limitations. (1) Calculating the accurate influence of all available data is time-consuming. (2) The selected data instances are not diverse enough, which may hinder the pretrained model's ability to generalize effectively to various downstream tasks. In this paper, we introduce $\texttt{Quad}$, a data selection approach that considers both quality and diversity by using data influence to achieve state-of-the-art pretraining results. To compute the influence ($i.e.,$ the quality) more accurately and efficiently, we incorporate the attention layers to capture more semantic details, which can be accelerated through the Kronecker product. For the diversity, $\texttt{Quad}$ clusters the dataset into similar data instances within each cluster and diverse instances across different clusters. For each cluster, if we opt to select data from it, we take some samples to evaluate the influence to prevent processing all instances. Overall, we favor clusters with highly influential instances (ensuring high quality) or clusters that have been selected less frequently (ensuring diversity), thereby well balancing between quality and diversity. Experiments on Slimpajama and FineWeb over 7B large language models demonstrate that $\texttt{Quad}$ significantly outperforms other data selection methods with a low FLOPs consumption. Further analysis also validates the effectiveness of our influence calculation.

Details

NeurIPS Conference 2025 Conference Paper

TANDEM: Bi-Level Data Mixture Optimization with Twin Networks

Jiaxing Wang
Deping Xiang
Jin Xu
Mingyang Yi
Guoqiang Gong
Zicheng Zhang
Haoran Li
Pengzhang Liu

The capabilities of large language models (LLMs) significantly depend on training data drawn from various domains. Optimizing domain-specific mixture ratios can be modeled as a bi-level optimization problem, which we simplify into a single-level penalized form and solve with twin networks: a proxy model trained on primary data and a dynamically updated reference model trained with additional data. Our proposed method, Twin Networks for bi-level DatA mixturE optiMization (TANDEM), measures the data efficacy through the difference between the twin models and up-weights domains that benefit more from the additional data. TANDEM provides theoretical guarantees and wider applicability, compared to prior approaches. Furthermore, our bi-level perspective suggests new settings to study domain reweighting such as data-restricted scenarios and supervised fine-tuning, where optimized mixture ratios significantly improve the performance. Extensive experiments validate TANDEM's effectiveness in all scenarios.

PDF Details

AAAI Conference 2024 Conference Paper

DINGO: Towards Diverse and Fine-Grained Instruction-Following Evaluation

Zihui Gu
Xingwu Sun
Fengzong Lian
Zhanhui Kang
Chengzhong Xu
Ju Fan

Instruction-following is particularly crucial for large language models (LLMs) to support diverse user requests. While existing work has made progress in aligning LLMs with human preferences, evaluating their capabilities on instruction-following remains a challenge due to complexity and diversity of real-world user instructions. While existing evaluation methods focus on general skills, they suffer from two main shortcomings, i.e., lack of fine-grained task-level evaluation and reliance on singular instruction expression. To address these problems, this paper introduces DINGO, a fine-grained and diverse instruction-following evaluation dataset that has two main advantages: (1) DINGO is based on a manual annotated, fine-grained and multi-level category tree with 130 nodes derived from real-world user requests; (2) DINGO includes diverse instructions, generated by both GPT-4 and human experts. Through extensive experiments, we demonstrate that DINGO can not only provide more challenging and comprehensive evaluation for LLMs, but also provide task-level fine-grained directions to further improve LLMs.

PDF Details DOI

AAAI Conference 2020 Conference Paper

Social Influence Does Matter: User Action Prediction for In-Feed Advertising

Hongyang Wang
Qingfei Meng
Ju Fan
Yuchen Li
Laizhong Cui
Xiaoman Zhao
Chong Peng
Gong Chen

Social in-feed advertising delivers ads that seamlessly ﬁt inside a user’s feed, and allows users to engage in social actions (likes or comments) with the ads. Many businesses pay higher attention to “engagement marketing” that maximizes social actions, as social actions can effectively promote brand awareness. This paper studies social action prediction for infeed advertising. Most existing works overlook the social in- ﬂuence as a user’s action may be affected by her friends’ actions. This paper introduces an end-to-end approach that leverages social inﬂuence for action prediction, and focuses on addressing the high sparsity challenge for in-feed ads. We propose to learn inﬂuence structure that models who tends to be inﬂuenced. We extract a subgraph with the near neighbors a user interacts with, and learn topological features of the subgraph by developing structure-aware graph encoding methods. We also introduce graph attention networks to learn inﬂuence dynamics that models how a user is inﬂuenced by neighbors’ actions. We conduct extensive experiments on real datasets from the commercial advertising platform of WeChat and a public dataset. The experimental results demonstrate that social inﬂuence learned by our approach can signiﬁcantly boost performance of social action prediction.

PDF Details

IJCAI Conference 2013 Conference Paper

Crowdsourcing-Assisted Query Structure Interpretation

Jun Han
Ju Fan
Lizhu Zhou

Structured Web search incorporating data from structured sources into search engine results has attracted much attention from both academic and industrial communities. To understand user’s intent, query structure interpretation is proposed to analyze the structure of queries in a query log and map query terms to the semantically relevant attributes of data sources in a target domain. Existing methods assume all queries should be classiﬁed to the target domain, and thus they are limited when interpreting queries from different domains in real query logs. To address the problem, we introduce a human-machine hybrid method by utilizing crowdsourcing platforms. Our method selects a small number of query terms and asks the crowdsourcing workers to interpret them, and then infers the interpretations based on the crowdsourcing results. To improve the performance, we propose an iterative probabilistic inference method based on a similarity graph of query terms, and select the most useful query terms for crowdsourcing by considering their domain-relevance and gained beneﬁt. We evaluate our method on a real query log, and the experimental results show that our method outperforms the state-of-the-art method.

PDF Details DOI