Papers by Zhi Xu

24 papers
VideoEraser: Concept Erasure in Text-to-Video Diffusion Models (2025.emnlp-main)

Copied to clipboard

Challenge: Experimental results show that VideoEraser outperforms prior methods regarding efficacy, integrity, fidelity, robustness, and generalizability.
Approach: They propose a training-free framework that prevents T2V diffusion models from generating videos with undesirable concepts even when explicitly prompted with those concepts.
Outcome: The proposed framework outperforms existing methods in erasure, celebrity erasion, and explicit content erasing tasks.
Exploring Schema Generalizability of Text-to-SQL (2023.findings-acl)

Copied to clipboard

Challenge: Existing text-to-SQL models are limited in their generalizability, despite their performance being over-estimated.
Approach: They propose a framework to generate novel text-to-SQL data via automatic and synchronous (DS, SQL) pair altering.
Outcome: The proposed framework generates text-to-SQL data via automatic and synchronous (DS, SQL) pair altering.
Granular Entity Mapper: Advancing Fine-grained Multimodal Named Entity Recognition and Grounding (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for fine-grained content extraction are limited by long-tailed distribution of textual entity categories and performance of object detectors.
Approach: They propose a multi-granularity entity recognition module and a reranking module to integrate hierarchical information of entity categories, visual cues, and external textual resources collectively.
Outcome: The proposed framework achieves state-of-the-art on the fine-grained content extraction task.
ShadowGNN: Graph Projection Neural Network for Text-to-SQL Parser (2021.naacl-main)

Copied to clipboard

Challenge: Existing semantic parsing models struggle to adapt to unseen database schemas . a new architecture, ShadowGNN, processes schemas at abstract and semantic levels .
Approach: They propose a new architecture which processes schemas at abstract and semantic levels.
Outcome: The proposed architecture outperforms state-of-the-art models on a text-to-sql benchmark . it uses domain-independent representations to extract logical linking between question and schema .
KAPA: A Deliberative Agent Framework with Tree-Structured Knowledge Base for Multi-Domain User Intent Understanding (2025.findings-acl)

Copied to clipboard

Challenge: Existing studies on the use of LLMs for estimating user intents are either too far from real human thought processes or require labeled samples.
Approach: They propose a deliberative agent framework that leverages human thought process to build high-level domain knowledge and a tree-structured knowledge base to store refined experience and data.
Outcome: The proposed framework is able to build high-level domain knowledge and efficiently store it across multiple steps.
VIGIL: Defending LLM Agents Against Tool-Stream Injection via Verify-Before-Commit (2026.acl-long)

Copied to clipboard

Challenge: Existing defenses for indirect prompt injection are limited by static protection mechanisms . existing models prioritize injected rules due to strict alignment, whereas static protections sever the feedback loop required for adaptive reasoning.
Approach: They propose a framework that shifts the paradigm from restrictive isolation to a verify-before-commit protocol.
Outcome: The proposed framework outperforms state-of-the-art dynamic defenses by reducing the attack success rate by over 22% while more thandoubling utility under attack compared to static baselines.
M3CoT: A Novel Benchmark for Multi-Domain Multi-step Multi-modal Chain-of-Thought (2024.acl-long)

Copied to clipboard

Challenge: MCoT requires models to leverage knowledge from both textual and visual modalities for step-by-step reasoning.
Approach: They propose a benchmark to address the challenges of MCoT, and evaluate it using vision large language models.
Outcome: The proposed benchmark addresses the above challenges and shows that current models still struggle to reason in M3CoT.
Rethinking the Role of Prompting Strategies in LLM Test-Time Scaling: A Perspective of Probability Theory (2025.acl-long)

Copied to clipboard

Challenge: Recent studies have shown that scaling test-time compute can also effectively improve reasoning.
Approach: They propose a probabilistic method to efficiently predict scaling performance and identify the best prompting strategy under large sampling times.
Outcome: The proposed method significantly improves the scaling performance of majority voting on large language models.
Assist Non-native Viewers: Multimodal Cross-Lingual Summarization for How2 Videos (2022.emnlp-main)

Copied to clipboard

Challenge: Existing multimodal summarization methods are limited to monolingual videos . a proposed task aims to generate cross-lingual summaries from multimodal inputs .
Approach: They propose a task to generate cross-lingual summaries from multimodal inputs of videos . they propose fusion network that integrates multimodal and cross-linguistic information .
Outcome: The proposed task outperforms existing methods on a reorganized How2 dataset on the reorganized How2 data set.
AMoPO: Adaptive Multi-objective Preference Optimization without Reward Models and Reference Models (2025.findings-acl)

Copied to clipboard

Challenge: Existing multi-objective preference alignment methods for large language models face limitations such as auxiliary reward/reference models and computational complexity.
Approach: They propose a framework that achieves dynamic balance across preference dimensions by using dimension-aware generation metrics as implicit rewards.
Outcome: Empirical results show that AMoPO outperforms state-of-the-art methods by 28.5% .
Double-Checker: Large Language Model as a Checker for Few-shot Named Entity Recognition (2024.findings-emnlp)

Copied to clipboard

Challenge: Recent studies have demonstrated remarkable performance on few-shot Named Entity Recognition tasks due to the high cost of obtaining high-quality labeled data.
Approach: They propose to decompose the task into entity span detection and entity type classification using a type-independent entity span detector and then classify the detected spans based on their types.
Outcome: The proposed method consistently yields improvements over two baseline approaches.
Towards Tool Use Alignment of Large Language Models (2024.emnlp-main)

Copied to clipboard

Challenge: Existing studies on tool use with LLMs focus on enhancing tool-calling ability of LLM . e.g., LLM should not answer unsafe tool use relevant instructions or insecure tool responses to ensure reliability and harmlessness.
Approach: They propose to use supervised fine-tuning and preference learning to align LLMs with H2A principle for tool use.
Outcome: The proposed model demonstrates that LLMs can generate truthful and helpful responses while remaining harmless.
MLeVLM: Improve Multi-level Progressive Capabilities based on Multimodal Large Language Model for Medical Visual Question Answering (2024.findings-acl)

Copied to clipboard

Challenge: Existing MVQA models ignore multi-level progressive capabilities due to unspecific data and plain architecture.
Approach: They propose a multi-level visual language model for medical visual question answering (MVQA) which covers multi- level questions and answers as well as reasoning processes from visual clues to semantic cognition.
Outcome: The proposed model outperforms existing medical multimodal large language models on a multi-level instruction dataset and a feature alignment module.
MLLM-I2W: Harnessing Multimodal Large Language Model for Zero-Shot Composed Image Retrieval (2025.coling-main)

Copied to clipboard

Challenge: Existing methods for combining image retrieval are supervised and zero-shot . however, the challenge of mapping pseudo-words to images within the joint image-text embedding space is still a challenge.
Approach: They propose a novel image-text mapping network which converts description-related image information into pseudo-word markers for precise ZS-CIR.
Outcome: The proposed model improves on COCO, CIRR, and Fashion-IQ benchmarks.
NoisyCausal: A Benchmark for Evaluating Causal Reasoning Under Structured Noise (2026.acl-long)

Copied to clipboard

Challenge: Existing benchmarks focus on clean, abstract scenarios where causal structure is simple or implicitly assumed.
Approach: They propose a benchmark to evaluate causal reasoning under structured noise.
Outcome: The proposed method outperforms standard prompting and reasoning baselines on NoisyCausal.
Following Occam’s Razor: Dynamic Combination of Structured Knowledge for Multi-Hop Question Answering using LLMs (2025.findings-emnlp)

Copied to clipboard

Challenge: Multi-hop question answering is a challenging task that requires capturing information from multiple positions in multiple documents.
Approach: They propose a framework for integrating text-based and triple-based paradigms that incorporates structured knowledge into large-scale question answering.
Outcome: The proposed framework improves multi-hop question answering by incorporating structured knowledge into the models.
Think and Recall: Layer-Level Prompting for Lifelong Model Editing (2025.emnlp-main)

Copied to clipboard

Challenge: Existing methods for lifelong model editing suffer from limitations in usability, such as requiring additional training corpora or lacking support for reversible and detachable edits.
Approach: They propose a plug-and-play method for knowledge retrieval and storage, i.e., Layer-Level Prompting, which enables seamless and efficient lifelong model editing.
Outcome: The proposed method outperforms existing methods on question answering and hallucination benchmarks across different LLMs.
Reasoning Over Space: Enabling Geographic Reasoning for LLM-Based Generative Next POI Recommendation (2026.acl-long)

Copied to clipboard

Challenge: Existing LLM-based recommenders lack explicit modeling of geographic signals . without explicit modeling geographic signals, recommenders struggle to capture core mobility patterns .
Approach: They propose a framework that utilizes geography as a decision variable within the reasoning process.
Outcome: The proposed framework achieves over 10% relative gains in hit rate over strongest LLM-based baselines and improves cross-city transfer.
Entropy-Aware Reshaping of Reinforcement Signals for Multi-Answer Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Reinforcement learning with verifiable rewards (RLVR) is a standard post-training paradigm for large language models.
Approach: They propose a framework that reshapes how learning signals are normalized and aggregated.
Outcome: Experiments on MCTACO and MMLU-Multi show that the proposed framework improves accuracy, training stability and cross-dataset transfer performance.
GUI0: Self-Evolving Foundational GUI Agents in Super App Ecosystems (2026.acl-long)

Copied to clipboard

Challenge: Automated interaction with graphical user interfaces (GUIs) is central to general artificial intelligence, but remains challenging within Super App ecosystems.
Approach: They propose a framework synergizing autonomous data synthesis with dual-agent co-evolution . GUI0 establishes a domain-aware foundation model via synthesized corpora and employs curriculum-driven reinforcement learning .
Outcome: The proposed framework outperforms Gemini-2.5-Pro and Claude-4-Sonnet in the SuperAPP benchmark and has universal efficacy across base models.
CodeDPO: Aligning Code Models with Self Generated and Verified Source Code (2025.acl-long)

Copied to clipboard

Challenge: Existing training methods for code generation do not improve code correctness and efficiency.
Approach: They propose a framework that integrates preference learning into code generation to improve code correctness and efficiency.
Outcome: The proposed framework improves code correctness and efficiency by integrating preference learning into code generation.
QueryAttack: Jailbreaking Aligned Large Language Models Using Structured Non-natural Query Language (2025.findings-acl)

Copied to clipboard

Challenge: Existing methods to bypass security defenses of large language models (LLMs) are not effective, but QueryAttack can be jailbroken.
Approach: They propose a framework to examine generalizability of safety alignment by translating malicious queries into structured non-natural query languages.
Outcome: The proposed framework can achieve high attack success rates and jailbreak various defense methods on mainstream LLMs.
Bridging Kernel Drivers and Virtual Device Models with LLM-Powered Automation (2026.acl-demo)

Copied to clipboard

Challenge: Linux kernel device drivers are tightly coupled with hardware, making them difficult to execute and test without physical devices.
Approach: They present a tool that generates QEMU-based virtual devices directly from Linux driver source code.
Outcome: The proposed tool generates QEMU-based virtual devices directly from Linux driver source code.
Memory-Augmented LLM-based Multi-Agent System for Automated Feature Generation on Tabular Data (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for automated feature generation rely on predefined operator libraries and do not incorporate feature semantics, limiting their ability to produce high-quality features.
Approach: They propose a Memory-Augmented LLM-based Multi-Agent System (MALMAS) that decomposes the generation process into agents with distinct responsibilities.
Outcome: The proposed method extracts informative features from raw tabular data without manual intervention and is crucial for accurate, generalizable machine learning.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations