Papers by Xiang Zeng

24 papers
Don’t Half-listen: Capturing Key-part Information in Continual Instruction Tuning (2025.acl-long)

Copied to clipboard

Challenge: Existing methods to improve instruction tuning for large language models may cause catastrophic forgetting (CF) CF is a problem where previously learned abilities are degraded .
Approach: They propose a continual instruction tuning method that uses key-part information gain to replay data and refine training objective.
Outcome: The proposed method achieves superior performance on both seen and held-out tasks.
Event-Radar: Event-driven Multi-View Learning for Multimodal Fake News Detection (2024.acl-long)

Copied to clipboard

Challenge: Existing methods for detecting multimedia fake news have demonstrated excellent results . however, addressing event-level inconsistency and learning from poor-quality news remains a challenge .
Approach: They propose an Event-diven fake news detection framework that integrates visual manipulation, textual emotion and multimodal inconsistency at event-level for fake news identification.
Outcome: The proposed framework performs well on three large-scale fake news detection benchmarks.
SafeSteer: A Decoding-level Defense Mechanism for Multimodal Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Existing defense methods rely on fine-tuning or inefficient post-hoc interventions, limiting their ability to address novel attacks.
Approach: They propose a decoding-level defense mechanism that employs a lightweight discriminator to iteratively steer the decoding process toward safety.
Outcome: The proposed method improves safety performance by up to 33.40% without fine-tuning on multiple MLLMs.
DRAMA: Dynamic Multi-Granularity Graph Estimate Retrieval over Tabular and Textual Question Answering (2024.lrec-main)

Copied to clipboard

Challenge: TableTextQA tasks require tabular and textual data, gaining increasing attention . however, row-based approaches suffer from limitations such as lack of interaction between rows .
Approach: They propose a method that incorporates an interaction mechanism among multiple rows . Empirical results demonstrate that the proposed method is effective .
Outcome: Empirical results show that the proposed model is effective on tabFact and HybridQA datasets.
Xiaomingbot: A Multilingual Robot News Reporter (2020.acl-demos)

Copied to clipboard

Challenge: Xiaomingbot is a multilingual and multimodal software robot with four capabilities: news generation, news translation, news reading and avatar animation.
Approach: They propose to build a multilingual and multimodal software robot with four inte- gal capabilities: news generation, news translation, news reading and avatar animation.
Outcome: The proposed system generates Chinese news, then reads it in multiple languages and generates an animated avatar reading it.
Automated Peer Reviewing in Paper SEA: Standardization, Evaluation, and Analysis (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing approaches to review scientific papers are limited by their content or quality . SEA is a framework for automated scientific review, but its contents are generic or partial.
Approach: They propose a framework for automated scientific review using large language models . they propose to use a standardized review dataset to fine-tune an LLM to generate high-quality reviews.
Outcome: The proposed framework can generate high-quality reviews from standardized datasets and improves on the existing feedback mechanisms.
Weakly Supervised Text Classification using Supervision Signals from a Language Model (2022.findings-naacl)

Copied to clipboard

Challenge: Existing weakly supervised text classification methods require a large number of annotated data and human annotations are expensive.
Approach: They propose to query a masked language model with cloze style prompts to obtain supervision signals.
Outcome: The proposed method outperforms baseline methods on three datasets by 2%, 4%, and 3%.
HiDe-LLaVA: Hierarchical Decoupling for Continual Instruction Tuning of Multimodal Large Language Model (2025.acl-long)

Copied to clipboard

Challenge: Existing methods to improve instructionfollowing performance of MLLMs often trade off memory efficiency for performance gains, compromising overall efficiency.
Approach: They propose a task-specific expansion and task-general fusion framework based on variations in Centered Kernel Alignment (CKA) similarity across different model layers when trained on diverse datasets.
Outcome: The proposed framework improves performance compared to existing benchmarks.
MixTEA: Semi-supervised Entity Alignment with Mixture Teaching (2023.findings-emnlp)

Copied to clipboard

Challenge: Existing methods to learn informative entity embeddings are insufficient for semi-supervised entity alignment.
Approach: They propose a semi-supervised method which guides the model learning with an end-to-end mixture teaching of manually labeled mappings and probabilistic pseudo mappings.
Outcome: The proposed method is superior to existing methods on benchmark datasets and further analyses.
HyperEdit: Unlocking Instruction-based Text Editing in LLMs via Hypernetworks (2026.findings-acl)

Copied to clipboard

Challenge: Existing approaches treat instruction-based text editing as a generic text generation problem. Existing methods either over-edit or fail to apply modifications consistently.
Approach: They propose a framework that processes each editing request to best align with it.
Outcome: The proposed framework achieves 9% improvement over the state-of-the-art model.
CLEEK: A Chinese Long-text Corpus for Entity Linking (2020.lrec-1)

Copied to clipboard

Challenge: Entity linking is a fundamental task in natural language processing, says nigel kilgstrom . existing corpora for entity linking in china are lacking and deficient, he says . kilsmstrom: a new method for entity disambiguation can be developed for Chinese .
Approach: They build a Chinese corpus of multi-domain long text for entity linking . they evaluate the difficulty of documents with respect to entity linking using a measure .
Outcome: The proposed corpus is based on 100 documents from diverse domains and is publicly accessible.
PromptSculptor: Multi-Agent Based Text-to-Image Prompt Optimization (2025.emnlp-demos)

Copied to clipboard

Challenge: PromptSculptor automates the iterative prompt optimization process for Text-to-Image models . previous work focused on generating detailed, high-quality prompts based on user feedback .
Approach: They propose a framework that decomposes a task into four specialized agents . they use Chain-of-Thought reasoning to transform a short, vague user prompt into a comprehensive, refined prompt.
Outcome: The proposed framework significantly improves output quality and reduces iterations needed for user satisfaction.
Stephanie: Step-by-Step Dialogues for Mimicking Human Interactions in Social Conversations (2025.findings-naacl)

Copied to clipboard

Challenge: a new paradigm for dialogue systems is being developed to mimic human interactions . the current single-step dialogue paradigm lacks the depth and fluidity of human interactions.
Approach: They propose a step-by-step dialogue paradigm that mimics human interactions . they use a dataset to fine-tune existing language models .
Outcome: The proposed system mimics the dynamic nature of human conversations . it is compared with existing paradigms and will be released later this year .
DPN-LE: Dual Personality Neuron Localization and Editing for Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Current methods for editing personality traits in large language models can change personalities but reduce performance.
Approach: They propose a novel paradigm for personality editing that locates and edits LLM neurons and enables competitive personality control at inference time.
Outcome: Experiments on LLaMA-3-8B-Instruct and Qwen2.5-7B-instruct show that the proposed approach can improve performance and improve performance.
Multi-Modal Entities Matter: Benchmarking Multi-Modal Entity Alignment (2025.coling-main)

Copied to clipboard

Challenge: Existing MMEA datasets consider multi-modal data as attributes of textual entities, neglecting correlations between the multi-modal data.
Approach: They propose a multi-modal entity alignment dataset that models multi-dimensional data as textual entities in the MMKG.
Outcome: The proposed dataset can learn the structural information of entities by considering both intra-modal and cross-modal relations and infer the similarity of different types of entity pairs.
Reinforcement Learning–Guided Adaptive Tuning for Out-of-Distribution Harmful Text Detection (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for testing harmful information on social media rely on fixed parameters that fail to handle substantial semantic discrepancies . RLAT can be used to adapt to semantic variations while preventing overfitting from continuous tuning.
Approach: They propose a reinforcement learning-guided adaptive tuning method for harmful text detection that optimizes consistency loss and applies word-level attention constraints to reduce over-reliance on local words.
Outcome: The proposed method outperforms state-of-the-art models in cross-platform and cross-temporal scenarios across multiple public datasets.
From Knowledge to Treatment: Large Language Model Assisted Biomedical Concept Representation for Drug Repurposing (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for drug repurposing ignore common-sense biomedical concept knowledge in real-world labs, such as mechanistic priors indicating that certain drugs are fundamentally incompatible with specific treatments.
Approach: They propose a Large Language Model-assisted framework for Drug Repurposing which improves the representation of biomedical concepts within KGs.
Outcome: The proposed framework improves the representation of biomedical concepts within KGs by extracting treatment-related textual representations of biomedic entities from large language models and fine-tuning knowledge graph embedding models.
RShield: A User-level Traceable Backdoor Watermark for LLMs in Embedding-as-a-Service (2026.findings-acl)

Copied to clipboard

Challenge: Existing backdoor watermarking techniques are limited to zero-bit detection . RShield enables reliable user-level attribution of large language models under model extraction attacks.
Approach: They propose a multi-bit backdoor watermarking technique that enables reliable user-level attribution of large language models under model extraction attacks.
Outcome: RShield achieves 100% multi-bit watermark recovery and high semantic fidelity under model extraction attacks compared to existing methods.
FARSS: Fisher-Optimized Adaptive Low-Rank and Singular-Vector Selection for Knowledge-Preserving Fine-Tuning (2026.findings-acl)

Copied to clipboard

Challenge: Low-rank adaptation methods for large language models have limitations in preserving world knowledge and limiting updates to preserve world knowledge.
Approach: They propose a Fisher-optimized adaptive low Rank and Singular-VectorSelection framework for knowledge-preserving fine-tuning that allows efficient and task-sensitive updates.
Outcome: The proposed framework outperforms existing methods for knowledge-preserving fine-tuning.
Unveiling Privacy Risks in LLM Agent Memory (2025.acl-long)

Copied to clipboard

Challenge: Large Language Model (LLM) agents store private user-agent interactions in memory for demonstrations, introducing new privacy risks for LLM agents.
Approach: They propose an attack that extracts private information from memory under a black-box setting and propose a method that can be used to attack the agent.
Outcome: The proposed attack is effective under a black-box setting and it is demonstrated on two representative agents.
DiffER: Diffusion Entity-Relation Modeling for Reversal Curse in Diffusion Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Existing large language models exhibit unidirectional behavior when processing bidirectional relationships . authors propose a solution to alleviate the reversal curse in Diffusion LLMs .
Approach: They propose a model that addresses the "reversal curse" of bidirectional behavior in large language models . they propose 'entity-aware training' and balanced data construction to alleviate asymmetry and missing relations .
Outcome: The proposed model alleviates the "reversal curse" in Diffusion LLMs . the proposed model employs whole-entity masking to mitigate entity fragmentation .
EFSA: Towards Event-Level Financial Sentiment Analysis (2024.acl-long)

Copied to clipboard

Challenge: a large-scale Chinese dataset contains 12,160 news articles and 13,725 quintuples . a four-hop Chain-of-Thought LLM-based approach is devised for this task .
Approach: They propose to extend financial sentiment analysis to event-level since events usually serve as the subject of the sentiment in financial text.
Outcome: The proposed method can reach the current state-of-the-art on a large-scale Chinese dataset.
OssCSE: Overcoming Surface Structure Bias in Contrastive Learning for Unsupervised Sentence Embedding (2023.emnlp-main)

Copied to clipboard

Challenge: Recent studies show that contrastive learning is effective in sentence representation learning . but, the surface structure bias is a problem in the current model .
Approach: They propose to combine a sentence with a sub-semantic sentence to investigate the surface structure bias.
Outcome: The proposed model achieves state-of-the-art on standard semantic textual similarity tasks using different pre-trained backbones.
KG-FiD: Infusing Knowledge Graph in Fusion-in-Decoder for Open-Domain Question Answering (2022.acl-long)

Copied to clipboard

Challenge: Open-Domain Question Answering (ODQA) models typically include a retrieving module and a reading module.
Approach: They propose a new open-domain question-answering framework that uses a knowledge-enhanced version of FiD to improve the approach.
Outcome: The proposed model improves on ODQA benchmark datasets with less than 40% computation cost.

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