Siyi Liu

Zhaonan Li, Shijie Lu, Fei Wang et al.

End-to-end Vision-language Models (VLMs) often answer visual questions by exploiting spurious correlations instead of causal visual evidence, and can become more shortcut-prone when fine-tuned. We introduce VISTA (Visual-Information Separation for Text-based Analysis), a modular framework that decouples perception from reasoning via an explicit information bottleneck. A frozen VLM sensor is restricted to short, objective perception queries, while a text-only LLM reasoner decomposes each question, plans queries, and aggregates visual facts in natural language. This controlled interface defines a reward-aligned environment for training unbiased visual reasoning with reinforcement learning. Instantiated with Qwen2.5-VL and Llama3.2-Vision sensors, and trained with GRPO from only 641 curated multi-step questions, VISTA significantly improves robustness to real-world spurious correlations on SpuriVerse (+16.29% with Qwen-2.5-VL-7B and +6.77% with Llama-3.2-Vision-11B), while remaining competitive on MMVP and a balanced SeedBench subset. VISTA transfers robustly across unseen VLM sensors and is able to recognize and recover from VLM perception failures. Human analysis further shows that VISTA's reasoning traces are more neutral, less reliant on spurious attributes, and more explicitly grounded in visual evidence than end-to-end VLM baselines.

Siyi Liu, Chen Gao, Yihong Chen et al.

The embedding-based representation learning is commonly used in deep learning recommendation models to map the raw sparse features to dense vectors. The traditional embedding manner that assigns a uniform size to all features has two issues. First, the numerous features inevitably lead to a gigantic embedding table that causes a high memory usage cost. Second, it is likely to cause the over-fitting problem for those features that do not require too large representation capacity. Existing works that try to address the problem always cause a significant drop in recommendation performance or suffers from the limitation of unaffordable training time cost. In this paper, we proposed a novel approach, named PEP (short for Plug-in Embedding Pruning), to reduce the size of the embedding table while avoiding the drop of recommendation accuracy. PEP prunes embedding parameter where the pruning threshold(s) can be adaptively learned from data. Therefore we can automatically obtain a mixed-dimension embedding-scheme by pruning redundant parameters for each feature. PEP is a general framework that can plug in various base recommendation models. Extensive experiments demonstrate it can efficiently cut down embedding parameters and boost the base model's performance. Specifically, it achieves strong recommendation performance while reducing 97-99% parameters. As for the computation cost, PEP only brings an additional 20-30% time cost compared with base models. Codes are available at https://github.com/ssui-liu/learnable-embed-sizes-for-RecSys.

Siyi Liu, Sihao Chen, Xander Uyttendaele et al.

We propose MultiOpEd, an open-domain news editorial corpus that supports various tasks pertaining to the argumentation structure in news editorials, focusing on automatic perspective discovery. News editorial is a genre of persuasive text, where the argumentation structure is usually implicit. However, the arguments presented in an editorial typically center around a concise, focused thesis, which we refer to as their perspective. MultiOpEd aims at supporting the study of multiple tasks relevant to automatic perspective discovery, where a system is expected to produce a single-sentence thesis statement summarizing the arguments presented. We argue that identifying and abstracting such natural language perspectives from editorials is a crucial step toward studying the implicit argumentation structure in news editorials. We first discuss the challenges and define a few conceptual tasks towards our goal. To demonstrate the utility of MultiOpEd and the induced tasks, we study the problem of perspective summarization in a multi-task learning setting, as a case study. We show that, with the induced tasks as auxiliary tasks, we can improve the quality of the perspective summary generated. We hope that MultiOpEd will be a useful resource for future studies on argumentation in the news editorial domain.

Yujia Zheng, Siyi Liu, Zekun Li et al.

This paper explores meta-learning in sequential recommendation to alleviate the item cold-start problem. Sequential recommendation aims to capture user's dynamic preferences based on historical behavior sequences and acts as a key component of most online recommendation scenarios. However, most previous methods have trouble recommending cold-start items, which are prevalent in those scenarios. As there is generally no side information in the setting of sequential recommendation task, previous cold-start methods could not be applied when only user-item interactions are available. Thus, we propose a Meta-learning-based Cold-Start Sequential Recommendation Framework, namely Mecos, to mitigate the item cold-start problem in sequential recommendation. This task is non-trivial as it targets at an important problem in a novel and challenging context. Mecos effectively extracts user preference from limited interactions and learns to match the target cold-start item with the potential user. Besides, our framework can be painlessly integrated with neural network-based models. Extensive experiments conducted on three real-world datasets verify the superiority of Mecos, with the average improvement up to 99%, 91%, and 70% in HR@10 over state-of-the-art baseline methods.

Yujia Zheng, Siyi Liu, Zekun Li et al.

The task of session-based recommendation is to predict user actions based on anonymous sessions. Recent research mainly models the target session as a sequence or a graph to capture item transitions within it, ignoring complex transitions between items in different sessions that have been generated by other users. These item transitions include potential collaborative information and reflect similar behavior patterns, which we assume may help with the recommendation for the target session. In this paper, we propose a novel method, namely Dual-channel Graph Transition Network (DGTN), to model item transitions within not only the target session but also the neighbor sessions. Specifically, we integrate the target session and its neighbor (similar) sessions into a single graph. Then the transition signals are explicitly injected into the embedding by channel-aware propagation. Experiments on real-world datasets demonstrate that DGTN outperforms other state-of-the-art methods. Further analysis verifies the rationality of dual-channel item transition modeling, suggesting a potential future direction for session-based recommendation.

Siyi Liu, Yujia Zheng

Session-based recommendation focuses on the prediction of user actions based on anonymous sessions and is a necessary method in the lack of user historical data. However, none of the existing session-based recommendation methods explicitly takes the long-tail recommendation into consideration, which plays an important role in improving the diversity of recommendation and producing the serendipity. As the distribution of items with long-tail is prevalent in session-based recommendation scenarios (e.g., e-commerce, music, and TV program recommendations), more attention should be put on the long-tail session-based recommendation. In this paper, we propose a novel network architecture, namely TailNet, to improve long-tail recommendation performance, while maintaining competitive accuracy performance compared with other methods. We start by classifying items into short-head (popular) and long-tail (niche) items based on click frequency. Then a novel is proposed and applied in TailNet to determine user preference between two types of items, so as to softly adjust and personalize recommendations. Extensive experiments on two real-world datasets verify the superiority of our method compared with state-of-the-art works.

Siyi Liu, Ziang Leng, Derry Wijaya

Mirroring is the behavior in which one person subconsciously imitates the gesture, speech pattern, or attitude of another. In conversations, mirroring often signals the speakers enjoyment and engagement in their communication. In chatbots, methods have been proposed to add personas to the chatbots and to train them to speak or to shift their dialogue style to that of the personas. However, they often require a large dataset consisting of dialogues of the target personalities to train. In this work, we explore a method that can learn to mirror the speaking styles of a person incrementally. Our method extracts ngrams that capture a persons speaking styles and uses the ngrams to create patterns for transforming sentences to the persons speaking styles. Our experiments show that our method is able to capture patterns of speaking style that can be used to transform regular sentences into sentences with the target style.

Yujia Zheng, Siyi Liu, Zailei Zhou

Predicting user actions based on anonymous sessions is a challenge to general recommendation systems because the lack of user profiles heavily limits data-driven models. Recently, session-based recommendation methods have achieved remarkable results in dealing with this task. However, the upper bound of performance can still be boosted through the innovative exploration of limited data. In this paper, we propose a novel method, namely Intra-and Inter-session Interaction-aware Graph-enhanced Network, to take inter-session item-level interactions into account. Different from existing intra-session item-level interactions and session-level collaborative information, our introduced data represents complex item-level interactions between different sessions. For mining the new data without breaking the equilibrium of the model between different interactions, we construct an intra-session graph and an inter-session graph for the current session. The former focuses on item-level interactions within a single session and the latter models those between items among neighborhood sessions. Then different approaches are employed to encode the information of two graphs according to different structures, and the generated latent vectors are combined to balance the model across different scopes. Experiments on real-world datasets verify that our method outperforms other state-of-the-art methods.