Papers by Xin Yuan

35 papers
Safe: Enhancing Mathematical Reasoning in Large Language Models via Retrospective Step-aware Formal Verification (2025.acl-long)

Copied to clipboard

Challenge: Chain-of-Thought prompting is a de facto method to elicit reasoning capabilities from large language models (LLMs).
Approach: They propose a step-aware formal verification framework Safe to address hallucinations in CoT prompting . they propose 'formal step' as a benchmark for step correctness theorem proving with 30,809 formal statements.
Outcome: The proposed framework shows significant performance improvement while offering interpretable and verifiable evidence.
ZSEE: A Dataset based on Zeolite Synthesis Event Extraction for Automated Synthesis Platform (2024.findings-naacl)

Copied to clipboard

Challenge: Automated synthesis of zeolite holds great significance for attaining economic and environmental benefits.
Approach: They propose an event extraction task to mine structural synthesis actions from experimental narratives for modular automated synthesis.
Outcome: The proposed method can significantly expedite automated synthesis of zeolites owing to its machine readability.
Chain of Ideas: Revolutionizing Research Via Novel Idea Development with LLM Agents (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for idea generation either trivially prompt LLMs or expose LLM to extensive literature without indicating useful information.
Approach: They propose a chain-of-ideas agent that organizes literature in a chains structure . they propose evaluating idea-generation methods from different perspectives .
Outcome: The proposed agent outperforms existing methods and matches human quality in idea generation.
AnyGPT: Unified Multimodal LLM with Discrete Sequence Modeling (2024.acl-long)

Copied to clipboard

Challenge: Existing language models that use discrete representations for unified processing of various modalities are limited to text generation and do not include multimodal output.
Approach: They propose a multimodal language model that utilizes discrete representations for unified processing of various modalities.
Outcome: The proposed model can be trained stably without any alterations to existing models or training paradigms.
SELECting over Tokens: Curating Pre-training Data at Scale via Token Classification (2026.acl-long)

Copied to clipboard

Challenge: Existing pipelines rely on expert-crafted heuristic rules, which lack content-aware, fine-grained noise detection.
Approach: They propose a framework that reframes data refinement as a highly efficient token classification task.
Outcome: The proposed framework outperforms existing pipelines on benchmarks and is 2.5x faster at inference.
Not The End of Story: An Evaluation of ChatGPT-Driven Vulnerability Description Mappings (2023.findings-acl)

Copied to clipboard

Challenge: Existing data proves that ChatGPT performs no less than humans in text generation and knowledge Q&A.
Approach: They propose to use ChatGPT to map vulnerabilities to common weakness enumeration (CWE), common attack pattern ennumeration and classification (ATT&CK) techniques and other classifications.
Outcome: The proposed method performs better than human experts on many tasks, but it can't replace professional security engineers in vulnerability analysis.
AgentCPM-GUI: Building Mobile-Use Agents with Reinforcement Fine-Tuning (2025.emnlp-demos)

Copied to clipboard

Challenge: Large language model agents have enabled GUI-based automation, but their deployment is limited by noisy data, poor generalization, and lack of support for non-English GUIs.
Approach: They propose an 8B-parameter GUI agent built for robust and efficient on-device GUI interaction.
Outcome: The proposed GUI agent achieves promising performance on five public benchmarks and proposed Chinese benchmark CAGUI.
SalaMAnder: Shapley-based Mathematical Expression Attribution and Metric for Chain-of-Thought Reasoning (2025.findings-emnlp)

Copied to clipboard

Challenge: Chain-of-Thought prompting improves the math reasoning capability of large language models.
Approach: They propose a method for attribution of component-level contributions in CoT reasoning using Shapley value and a stratified sampling algorithm that significantly reduces computational complexity.
Outcome: The proposed method reduces computational complexity and provides robust correlations with model performance.
BERT-MK: Integrating Graph Contextualized Knowledge into Pre-trained Language Models (2020.findings-emnlp)

Copied to clipboard

Challenge: Existing knowledge representation learning methods do not use graph contextualized knowledge.
Approach: They propose to model subgraphs in a medical KG and integrate it with a pre-trained language model to do knowledge generalization.
Outcome: The proposed model achieves state-of-the-art on several medical NLP tasks . it improves on MedERNIE, and the proposed model is effective .
DGPO: Beyond Pairwise Preferences with Directional Consistent Groupwise Optimization (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for directional consistency alignment of large language models are limited . a recent study suggests reverse supervision as a complement to forward reasoning .
Approach: They propose a framework that aggregates supervision signals at the group level and explicitly models direction-aware alignment through multi-candidate comparisons.
Outcome: The proposed framework achieves 3.2% accuracy improvement across five benchmarks and multiple datasets.
Support or Refute: Analyzing the Stance of Evidence to Detect Out-of-Context Mis- and Disinformation (2023.emnlp-main)

Copied to clipboard

Challenge: Mis- and disinformation online are a major source of harms of different kinds . out-of-context information is where different pieces of information are falsely associated . past studies have attempted to defend against OOC mis- and deinformation through external evidence, but they disregard the role of different pieces with different stances.
Approach: They propose a stance extraction network that can extract stances of different pieces of evidence in a single framework.
Outcome: The proposed model outperforms the state-of-the-art models on a public large-scale dataset with a performance gain of 3.2% in accuracy.
VisCGEC: Benchmarking the Visual Chinese Grammatical Error Correction (2025.naacl-long)

Copied to clipboard

Challenge: Existing studies on Chinese grammatical error correction ignore multi-modality and faked errors, which pushes techniques far away from real-world scenarios.
Approach: They propose to benchmark Chinese grammatical error correction for Chinese as a foreign language learner (CFL) using a dataset, they propose to use two CGEC frameworks to conduct experiments .
Outcome: The proposed approach achieves an F 0.5 score of only 28.9%.
Diffusion Based Counterfactual Augmentation for Dual Sentiment Classification (2024.lrec-main)

Copied to clipboard

Challenge: Existing methods to augment sentiment models have failed to mitigate spurious association problem inherent in the original data.
Approach: They propose a framework for enhancing sentiment models using an antonymous paradigm and contrastive learning to generate high-quality samples.
Outcome: The proposed framework achieves state-of-the-art performance on four benchmark datasets.
Batch IS NOT Heavy: Learning Word Representations From All Samples (P18-1)

Copied to clipboard

Challenge: Stochastic Gradient Descent with negative sampling is the most prevalent approach to learn word representations.
Approach: They propose a method that uses batch gradient learning to generate word representations from all training samples.
Outcome: The proposed method outperforms sampling-based methods on several benchmark tasks.
ShortGPT: Layers in Large Language Models are More Redundant Than You Expect (2025.findings-acl)

Copied to clipboard

Challenge: Recent studies have identified significant redundancy in large language models . quantization and pruning are two methods that reduce computational resources .
Approach: They propose simple pruning methods that prune redundant layers based on their BI scores.
Outcome: The proposed pruning methods demonstrate superior performance over previous pruning methods.
Truth or Sophistry? LoFa: A Benchmark for LLM Robustness Against Logical Fallacies (2026.acl-long)

Copied to clipboard

Challenge: Prior work has focused on the ability of Large Language Models to **identify** or **classify** fallacies, but their robustness against these fallacias in persuasive contexts remains largely unexplored.
Approach: They propose a new metric to assess LLM robustness against fallacies by pairing factual questions with fallacious arguments and developing a multi-round debate framework to assess model resilience.
Outcome: The proposed metric disentangles robustness from a model’s knowledge limitations and demonstrates unique vulnerability profiles across models.
NaviMaster: Learning a Unified Policy for GUI and Embodied Navigation Tasks (2026.acl-long)

Copied to clipboard

Challenge: Recent advances in Graphical User Interface (GUI) and embodied navigation have driven progress, yet these domains have largely evolved in isolation, with disparate datasets and training paradigms.
Approach: They propose a visual-target trajectory collection pipeline that generates trajectories for GUI and embodied tasks using a single formulation.
Outcome: The proposed agent outperforms state-of-the-art agents in GUI navigation, spatial affordance prediction, and embodied navigation.
Global Context or Local Detail? Adaptive Visual Grounding for Hallucination Mitigation (2026.findings-acl)

Copied to clipboard

Challenge: Large vision–language models suffer from object-existence hallucinations when multi-step deliberation decouples from visual evidence.
Approach: They propose a framework that allocates visual computation by uncertainty . they propose highlighting retains global context, while selective zoom-in performs local verification.
Outcome: The proposed framework reduces the complexity of multimodal reasoning by minimizing the operator trade-off.
pFedRAG: A Personalized Federated Retrieval-Augmented Generation System with Depth-Adaptive Tiered Embedding Tuning (2025.findings-emnlp)

Copied to clipboard

Challenge: Personalized Federated RAG framework enables efficient collaborative fine-tuning of embedding models . depth-adaptive tieered Embedding (DATE) architecture is tailored for local data and training results of each client.
Approach: a new Personalized Federated RAG framework is proposed for large language models . the framework enables efficient collaborative fine-tuning of embedding models based on common knowledge .
Outcome: a novel Personalized Federated RAG framework is proposed for large language models . the framework enables efficient collaborative fine-tuning of embedding models based on common knowledge .
PretrainRL: Alleviating Factuality Hallucination of Large Language Models at the Beginning (2026.findings-acl)

Copied to clipboard

Challenge: Large language models suffer from factual hallucinations where they generate verifiable falsehoods.
Approach: They propose a framework that integrates reinforcement learning into the pretraining phase to consolidate factual knowledge.
Outcome: The proposed framework significantly alleviates factual hallucinations and outperforms state-of-the-art methods.
Jailbreaking Attacks vs. Content Safety Filters: How Far Are We in the LLM Safety Arms Race? (2026.findings-acl)

Copied to clipboard

Challenge: Existing studies have focused on the models, neglecting the full deployment pipeline . previous studies have underestimated the practical success of these attacks .
Approach: They evaluate the effectiveness of jailbreak attacks targeting LLM safety alignment . they highlight critical gaps and call for further refinement of detection accuracy and usability .
Outcome: The proposed attacks can detect at least one safety filter across the entire deployment pipeline.
An Effective and Efficient Entity Alignment Decoding Algorithm via Third-Order Tensor Isomorphism (2022.acl-long)

Copied to clipboard

Challenge: Existing methods focus on graph representation learning, but decoding is a key part of the process.
Approach: They propose an EA Decoding Algorithm via Third-order Tensor Isomorphism (DATTI) they combine two sets of isomorphic equations to enhance the decoding process .
Outcome: The proposed algorithm can deliver significant performance improvements even on the most advanced methods while the extra required time is less than 3 seconds.
HydraRAG: Structured Cross-Source Enhanced Large Language Model Reasoning (2025.emnlp-main)

Copied to clipboard

Challenge: Current RAG system retrieves evidence from knowledge graphs and text documents but has limitations in multi-hop reasoning, multi-entity questions, and source verification.
Approach: They propose a training-free framework that unifies graph topology, document semantics, and source reliability to support deep, faithful reasoning in large language models.
Outcome: The proposed framework outperforms the current hybrid model-based model-driven system by 20.3% and 30.1% on seven benchmark datasets.
Mapping Natural Language Instructions to Mobile UI Action Sequences (2020.acl-main)

Copied to clipboard

Challenge: a new problem of grounding natural language instructions to mobile UI actions is emerging . we use a Transformer to extract action phrase tuples from long-range natural language instruction .
Approach: They propose a dataset that pairs English instructions with actions performed by people on a mobile UI emulator.
Outcome: The proposed model achieves 70.59% accuracy on predicting complete ground-truth action sequences in PixelHelp.
SolEval: Benchmarking Large Language Models for Repository-level Solidity Smart Contract Generation (2025.emnlp-main)

Copied to clipboard

Challenge: Existing methods focus on Python and Java, neglecting Solidity, the programming language for Ethereum smart contracts.
Approach: They construct a repository-level benchmark for Solidity to evaluate the performance of LLMs on Ethereum.
Outcome: The proposed benchmarks show that the best performing LLM achieves only 26.29% Pass@10, highlighting room for improvement in Solidity code generation.
SAFE-QAQ: End-to-End Slow-Thinking Audio-Text Fraud Detection via Reinforcement Learning (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for fraud detection rely on transcribed text, lacking acoustic cues . a proposed framework for audio-based slow-thinking fraud detection eliminates transcription errors .
Approach: They propose a framework for audio-based slow-thinking fraud detection that eliminates transcription errors and rewards slow-thought reasoning by capturing fine-grained audio details.
Outcome: The proposed method improves accuracy, inference efficiency, and real-time processing capabilities.
CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark (2022.acl-long)

Copied to clipboard

Challenge: a new benchmark for biomedical language understanding is being developed in Chinese . most benchmarks are limited to English, which makes it difficult to replicate success in other languages.
Approach: They propose to use Chinese biomedical language understanding evaluation benchmarks to evaluate Chinese models.
Outcome: The proposed benchmarks show that the current models perform worse than the human ceiling.
Improving Zero-Shot Entity Linking Candidate Generation with Ultra-Fine Entity Type Information (2022.coling-1)

Copied to clipboard

Challenge: Entity linking is a task of assigning entity mentions to referent entities in a knowledge base.
Approach: They propose to use ultra-fine-grained type information to improve the generalization ability of EL models by utilizing a low-level task to extract ultra-finish entity type information.
Outcome: The proposed model achieves state-of-the-art in the zero-shot entity linking task .
PrivLM-Bench: A Multi-level Privacy Evaluation Benchmark for Language Models (2024.acl-long)

Copied to clipboard

Challenge: generative large language models (LLMs) exhibit surprising capability and integrate previous tasks into a unified text generation formulation.
Approach: They propose a privacy evaluation benchmark to quantify the privacy leakage of language models.
Outcome: The proposed benchmark compares PPLMs with different privacy implementations to find out how privacy leakage is handled.
Aspect Sentiment Quad Prediction as Paraphrase Generation (2021.emnlp-main)

Copied to clipboard

Challenge: Existing studies focus on predicting the four elements in one shot, instead of predicting them all.
Approach: They propose a task to jointly detect all sentiment elements in quads for a given opinionated sentence.
Outcome: The proposed method can generate the semantics of the sentiment elements in the natural language form.
A Simple Temporal Information Matching Mechanism for Entity Alignment between Temporal Knowledge Graphs (2022.coling-1)

Copied to clipboard

Challenge: Existing methods for EA between temporal KGs incorporate relational and temporal information into entity embeddings.
Approach: They propose a method to generate unsupervised alignment seeds using temporal information from TKGs.
Outcome: The proposed method outperforms the previous methods by using temporal information.
TextFlint: Unified Multilingual Robustness Evaluation Toolkit for Natural Language Processing (2021.acl-demo)

Copied to clipboard

Challenge: Existing approaches to textual robustness evaluation focus on slightly modifying the input data, which maintains the original meaning and results in a different prediction.
Approach: They propose a multilingual robustness evaluation toolkit for NLP that integrates universal text transformations, task-specific transformations and adversarial attack.
Outcome: The toolkit includes universal text transformation, task-specific transformation, adversarial attack, subpopulation, and their combinations to provide comprehensive robustness analyses.
DialogVCS: Robust Natural Language Understanding in Dialogue System Upgrade (2024.naacl-long)

Copied to clipboard

Challenge: Existing models for natural language understanding are based on a well-defined intent 1 ontology.
Approach: They propose to retrain the natural language understanding model as new data from real users are merged into existing data.
Outcome: The proposed model shows that the semantically entangled intents can be recognized with an automatic workflow.
Unleashing Low-Bit Inference on Ascend NPUs: A Comprehensive Evaluation of HiFloat Formats (2026.acl-industry)

Copied to clipboard

Challenge: Low-bit floating-point formats like MXFP and NVFP4 offer new opportunities for precision and efficiency.
Approach: They evaluate HiFloat (HiF8 and HiF4), a family of floating-point formats tailored for Ascend NPUs.
Outcome: The proposed formats excel with high-variance data and are compatible with state-of-the-art quantization frameworks.
Towards Unified Representations of Knowledge Graph and Expert Rules for Machine Learning and Reasoning (2022.aacl-main)

Copied to clipboard

Challenge: Empirical study shows superiority of proposed method over time-tested knowledge-driven and data-driven methods.
Approach: They propose a cognitive knowledge graph that unifies expert rules and relational facts as the substrate of machine learning and reasoning models.
Outcome: Empirical results show the proposed method superior to time-tested methods . the proposed model can perform both learning and reasoning with labeled data .

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