Author name cluster

Xinxin Liu

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.

4 papers

1 author row

TCS Journal 2026 Journal Article

Image reflection on process graphs of 1-free regular expressions modulo bisimilarity

Yuanrui Zhang
Xinxin Liu

We study a phenomenon called “image reflection” on a type of characterization graphs — LLEE charts — for 1-free regular expressions modulo bisimularity. This property, stating that the images of a bisimulation function from an LLEE chart actually impose a special LEE structure corresponding to the LLEE chart, is recognized by our proposed “well-structured looping-back charts” as a sub-LLEE-structure of LLEE charts. As an application, our study naturally leads to a novel proof for the completeness of the inference system BBP for 1-free regular expressions, due to the correspondence between 1-free regular expressions and the provable solutions of LEE/LLEE charts. Compared to the previous approach, our proof is more direct in the sense that it does not rely on a graph transformation procedure on LLEE charts in which at each step two bisimilar nodes have to be carefully selected and merged together according to selection rules. Our observation on LLEE charts is useful to understand the completeness problems of regular expressions modulo bisimilarity from a new angle, and can be also helpful for solving the completeness problems of other expressions that share similar graph structures.

Details DOI

AAAI Conference 2026 Conference Paper

LaTeX2Layout: High-Fidelity, Scalable Document Layout Annotation Pipeline for Layout Detection

Feijiang Han
Zelong Wang
Bowen Wang
Xinxin Liu
Skyler Cheung
Delip Rao
Chris Callison-Burch
Lyle Ungar

General-purpose Vision-Language Models (VLMs) are increasingly integral to modern AI systems for document understanding, yet their ability to perform fine-grained layout analysis remains severely underdeveloped. Overcoming this limitation requires large-scale, high-fidelity training datasets. However, current annotation methods that rely on parsing rendered PDFs are costly, error-prone, and difficult to scale. We propose a different paradigm: extracting ground-truth layout directly from the LaTeX compilation process rather than the final PDF. We present LaTeX2Layout, a generalizable procedural pipeline that recovers pixel-accurate bounding boxes and reading order from compiler traces. This enables the generation of a 140K-page dataset, including 120K programmatically generated synthetic variants that more than double the layout diversity of real-world data. Using this dataset, we fine-tune an efficient 3B-parameter VLM with an easy-to-hard curriculum that accelerates convergence. Our model achieves Kendall's tau=0.95 for reading order and mAP@50=0.91 for element grounding, delivering nearly 200% relative improvement over strong zero-shot baselines such as GPT-4o and Claude-3.7.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

CPO: Condition Preference Optimization for Controllable Image Generation

Zonglin Lyu
Ming Li
Xinxin Liu
Chen Chen

To enhance controllability in text-to-image generation, ControlNet introduces image-based control signals, while ControlNet++ improves pixel-level cycle consistency between generated images and the input control signal. To avoid the prohibitive cost of back-propagating through the sampling process, ControlNet++ optimizes only low-noise timesteps (e. g. , $t < 200$) using a single-step approximation, which not only ignores the contribution of high-noise timesteps but also introduces additional approximation errors. A straightforward alternative for optimizing controllability across all timesteps is Direct Preference Optimization (DPO), a fine-tuning method that increases model preference for more controllable images ($I^{w}$) over less controllable ones ($I^{l}$). However, due to uncertainty in generative models, it is difficult to ensure that win--lose image pairs differ only in controllability while keeping other factors, such as image quality, fixed. To address this, we propose performing preference learning over control conditions rather than generated images. Specifically, we construct winning and losing control signals, $\mathbf{c}^{w}$ and $\mathbf{c}^{l}$, and train the model to prefer $\mathbf{c}^{w}$. This method, which we term \textit{Condition Preference Optimization} (CPO), eliminates confounding factors and yields a low-variance training objective. Our approach theoretically exhibits lower contrastive loss variance than DPO and empirically achieves superior results. Moreover, CPO requires less computation and storage for dataset curation. Extensive experiments show that CPO significantly improves controllability over the state-of-the-art ControlNet++ across multiple control types: over $10\%$ error rate reduction in segmentation, $70$--$80\%$ in human pose, and consistent $2$--$5\%$ reductions in edge and depth maps. The error rate is defined as the difference between the evaluated controllability and the oracle results. Our project is available \textcolor{blue}{\href{https: //zonglinl. github. io/CPO_page}{here}}.

PDF Details

TCS Journal 1998 Journal Article

Partial confluence of processes and systems of objects

Xinxin Liu
David Walker

A process calculus extending the π-calculus with higher-order agent abstractions as in the Higher-Order π-calculus and first-order data other than names but with only first-order interaction is used to give a natural and direct semantic definition for a concurrent object-oriented programming language. A notion of partial confluence of agents is introduced and its theory developed, first in the setting of CCS and then in the mobile-process calculus. It is shown how the semantic definition can be used as a basis for reasoning about systems prescribed by programs of the language: the theory of partial confluence is used to prove the indistinguishability in an arbitrary program context of two classes whose instances combine to form data structures only one of which supports concurrent operations.

Details DOI