Papers by Zirui Zhang

25 papers
The Stepwise Deception: Simulating the Evolution from True News to Fake News with LLM Agents (2025.emnlp-main)

Copied to clipboard

Challenge: Existing studies assume fake news is inherently existing rather than exploring its gradual formation.
Approach: They propose a Large Language Model-based simulation approach explicitly focusing on fake news evolution from real news.
Outcome: The proposed framework captures fake news evolution patterns and accurately reproduces known fake news, aligning closely with human evaluations.
ClusterAttn: KV Cache Compression under Intrinsic Attention Clustering (2025.acl-long)

Copied to clipboard

Challenge: Existing methods for sparse attention apply the same pattern across different attention heads and inputs, but fail to capture the intrinsic attention clustering in large language models.
Approach: They propose a training-free sparse attention method that provides an efficient prompt cache compression scheme under intrinsic attention clustering for efficient LLM inference.
Outcome: The proposed method reduces memory usage by 10%–65% and increases throughput by 2.6–4.8 times with no accuracy loss.
ServImage: An Image Generation and Editing Benchmark from Real-world Commercial Imaging Services (2026.acl-long)

Copied to clipboard

Challenge: Recent image generation and editing models demonstrate robust adherence to instructions and high visual quality on academic benchmarks.
Approach: They propose a benchmark that correlates image outputs with economic value in commercial design projects.
Outcome: ServImage benchmarks show image generation models perform well on academic benchmarks but are uncertain on commercial projects.
Towards Human-aligned Evaluation for Linear Programming Word Problems (2024.lrec-main)

Copied to clipboard

Challenge: Existing evaluation methodologies for MWPs diverge from human judgment and face challenges in recognizing mathematically equivalent answers.
Approach: They propose an evaluation metric rooted in graph edit distance that features benefits such as permutation invariance and more accurate program equivalence identification.
Outcome: The proposed evaluation metric features benefits such as permutation invariance and more accurate program equivalence identification.
UnifEE: Unified Evidence Extraction for Fact Verification (2023.eacl-main)

Copied to clipboard

Challenge: Existing models extract evidence in both sentences and table cells from Wikipedia dumps, ignoring potential connections between them.
Approach: They propose a model which uses a mixed evidence graph to extract the evidence in both formats without manually designed conversion rules.
Outcome: The proposed model outperforms existing models and improves the verification step.
Under the Shadow of Babel: How Language Shapes Reasoning in LLMs (2025.findings-emnlp)

Copied to clipboard

Challenge: linguistic relativity suggests that the structure of language shapes cognitive patterns . large language models internalize the habitual logical structures embedded in different languages, authors say .
Approach: a study introduces a bilingual dataset for causal reasoning in Chinese and English.
Outcome: a new study shows that large language models internalize reasoning biases shaped by language . the model internalizes language-specific preferences and rigidly applies them to atypical inputs, the study shows .
Audio Jailbreak: An Open Comprehensive Benchmark for Jailbreaking Large Audio-Language Models (2026.acl-long)

Copied to clipboard

Challenge: a recent study evaluated large audio-language models against jailbreak attacks . a new benchmark is being developed to evaluate LAM safety against jailbreaking attacks based on temporal and semantic nature of speech .
Approach: They propose a benchmark to evaluate LAM jailbreak vulnerabilities in adversarial audio prompts . they use a dataset of 1,495 adversarials to evaluate their performance .
Outcome: The proposed benchmark evaluates state-of-the-art LAMs against jailbreak attacks . it demonstrates that even small, semantically preserved perturbations can reduce safety .
SAFER: A Controllable Safeguard for LLMs against Backdoor Attacks (2026.findings-acl)

Copied to clipboard

Challenge: Existing inference-time defenses lack explicit control over false acceptance rate (FAR) existing inference time defenses aim to mitigate poisoned inputs but lack explicit FAR control .
Approach: They propose a framework that provides explicit control over false acceptance rate without prior knowledge of backdoor samples.
Outcome: The proposed framework outperforms existing inference-time defenses on three benchmark datasets . it provides explicit and provable control over false acceptance rate without prior knowledge of backdoor samples .
ARM: An Alignment-and-Replacement Module for Chinese Spelling Check Based on LLMs (2024.emnlp-main)

Copied to clipboard

Challenge: Chinese Spelling Check (CSC) aims to identify and correct spelling errors in Chinese texts, where enhanced semantic understanding of a sentence can significantly improve correction accuracy.
Approach: They propose a plug-and-play Alignment-and -Replacement module that enhances existing Chinese CSC models without retraining or fine-tuning.
Outcome: The proposed module improves existing models while reducing retraining and fine-tuning.
Augmenting Operations Research with Auto-Formulation of Optimization Models From Problem Descriptions (2022.emnlp-industry)

Copied to clipboard

Challenge: Existing systems for operations research use NLP to suggest formulations of optimization problems.
Approach: They propose an augmented intelligence system that can be used to simplify and enhance the modeling experience for operations research.
Outcome: The proposed system validates and edits the proposed formulations with a dataset of linear programming problems drawn from various application domains.
MMAU: A Holistic Benchmark of Agent Capabilities Across Diverse Domains (2025.findings-naacl)

Copied to clipboard

Challenge: Existing benchmarks focus on specific application scenarios, emphasizing task completion but failing to dissect the underlying skills that drive these outcomes.
Approach: They propose a Massive Multitask Agent Understanding benchmark that evaluates LLMs across five domains and offline tasks.
Outcome: The Massive Multitask Agent Understanding (MMAU) benchmark evaluates models across five domains including Tool-use, Directed Acyclic Graph (DAG) QA, Data Science and Machine Learning coding, Contest-level programming and Mathematics.
SAE-FiRE: Enhancing Earnings Surprise Predictions Through Sparse Autoencoder Feature Selection (2026.findings-acl)

Copied to clipboard

Challenge: Conference call transcripts contain significant redundancy and industry-specific terminology that creates obstacles for language models.
Approach: They propose a Sparse Autoencoder for Financial Representation Enhancement framework to extract key information from earnings conference call transcripts and eliminate redundancy.
Outcome: The proposed method outperforms baselines in analyzing earnings conference call transcripts.
Controllable Contamination Detection for Reliable LLM Evaluation with Statistical Guarantees (2026.acl-long)

Copied to clipboard

Challenge: Existing training data detectors fail to detect clean samples from contaminated test sets . existing methods fail to identify clean samples due to black-box nature of LLMs .
Approach: They propose a framework that detects and filters contaminated evaluation data . they propose 'failure detection' to reduce the proportion of contaminated samples mistakenly retained .
Outcome: The proposed framework reduces false discovery rate (FDR) under valid FDR control while maintaining evaluation consistency.
DSQG-Syn: Synthesizing High-quality Data for Text-to-SQL Parsing by Domain Specific Question Generation (2025.findings-naacl)

Copied to clipboard

Challenge: Existing methods for generating SQL queries using natural language questions produce inconsistent NLQ-SQL pairs.
Approach: They propose a text-to-SQL data synthesis framework that generates domain-relevant questions . they synthesize NLQ-SqL pairs that are domain-specific and intent-consistent .
Outcome: The proposed method outperforms closed-source LLMs on the Text-to-SQL task.
ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities (2025.findings-naacl)

Copied to clipboard

Challenge: Recent advances in large language models have led to a growing interest in tool assisted LLMs . toolSandbox includes stateful tool execution, implicit state dependencies between tools .
Approach: a new tool-based evaluation tool is released to help LLMs evaluate their tool-use capabilities. a tool-driven evaluation tool includes stateful tool execution, implicit state dependencies between tools and a built-in user simulator.
Outcome: the toolSandbox evaluation benchmark shows that open source and proprietary models have a performance gap . the benchmarks show that even the most capable LLMs are challenged by state dependent tasks .
Hazards in Daily Life? Enabling Robots to Proactively Detect and Resolve Anomalies (2025.naacl-long)

Copied to clipboard

Challenge: Existing household robots are inability to recognize potential problems or dangers in home environments.
Approach: They propose a task of creating anomaly scenarios using generative models instead of manually labeled data to build simulated environments.
Outcome: The proposed framework outperforms existing models in terms of task description and scene diversity.
When Personalization Tricks Detectors: The Feature-Inversion Trap in Machine-Generated Text Detection (2026.acl-long)

Copied to clipboard

Challenge: Personalized MGT detection remains largely underexplored due to personalization challenges . large language models (LLMs) can imitate personal writing styles, but they can generate fake news and misinformation.
Approach: They propose a benchmark to evaluate detector robustness under personalization . they attribute this limitation to a feature-inversion trap that flips the effect in personalized contexts .
Outcome: The proposed framework predicts detector robustness under personalization with an 85% correlation to actual results.
AEGIS: A Holistic Benchmark for Evaluating Forensic Analysis of AI-Generated Academic Images (2026.acl-long)

Copied to clipboard

Challenge: AEGIS examines whether current models can effectively audit AI-generated images in academic papers.
Approach: They propose a holistic benchmark for forensic analysis of AI-Generated academic ImageS that reveals limitations in academic image forensics.
Outcome: AEGIS compared with existing benchmarks on seven academic categories and features key advances in forensic analysis.
SAE-SSV: Supervised Steering in Sparse Representation Spaces for Reliable Control of Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) have impressive capabilities in natural language understanding and generation, but controlling their behavior remains a challenge.
Approach: They propose a supervised steering approach that operates in sparse, interpretable representation spaces.
Outcome: The proposed approach achieves higher success rates with minimal degradation in generation quality compared to existing methods.
MedINST: Meta Dataset of Biomedical Instructions (2024.findings-emnlp)

Copied to clipboard

Challenge: Medical data and tasks require extensive preprocessing and standardization for effective use in training LLMs.
Approach: They propose to use MedINST as a meta-dataset to evaluate LLMs' generalization ability.
Outcome: The meta-dataset of biomedical instruction measures the generalization ability of LLMs across multiple open-domain tasks.
Prompt Tuning for Few-shot Relation Extraction via Modeling Global and Local Graphs (2024.lrec-main)

Copied to clipboard

Challenge: Recent studies show that prompt-tuning is effective for few-shot relation extraction tasks.
Approach: They propose to incorporate the knowledge in relation labels into prompt-tuning by inserting prompt templates into the input.
Outcome: The proposed method improves on four datasets under low-resource conditions.
Choosing Transfer Languages for Cross-Lingual Learning (P19-1)

Copied to clipboard

Challenge: Cross-lingual transfer is a useful tool for improving performance of natural language processing (NLP) on low-resource languages.
Approach: They propose to use cross-lingual transfer to improve accuracy of low-resource languages . they build models that consider features to perform prediction on such languages based on ranking problem .
Outcome: The proposed model predicts good transfer languages much better than baselines considering single features in isolation.
Seek-and-Solve: Benchmarking MLLMs for Visual Clue-Driven Reasoning in Daily Scenarios (2026.findings-acl)

Copied to clipboard

Challenge: Existing benchmarks focus on evaluating MLLMs’ pre-existing knowledge or perceptual understanding, often neglecting the critical capability of reasoning.
Approach: They propose a benchmark designed for visual clue-driven reasoning in daily scenarios that combines rigorous grounding in authentic daily activities and challenging query design that necessitates more than surface-level perception.
Outcome: The proposed benchmark identifies visual clues and their ability to provide robust reasoning in daily scenarios.
Example Quality Matters: Multi-Aspects Example Augmentation for Private Library Programming (2026.acl-long)

Copied to clipboard

Challenge: Existing approaches to code generation fail to consider the quality of retrieved examples.
Approach: They propose a retrieval-augmented generation method that combines existing API examples to improve complexity and readability.
Outcome: The proposed method achieves up to 22% accuracy improvement over baseline methods.

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