Papers by Xue Yang

SafeSteer: A Decoding-level Defense Mechanism for Multimodal Large Language Models (2026.findings-acl)

Challenge:	Existing work regards dropped pronoun recovery and conversational discourse parsing as two separate tasks and tackles them separately.
Approach:	They propose a neural model for dropped pronoun recovery and conversational discourse parsing in Chinese conversational speech.
Outcome:	The proposed model outperforms the state-of-the-art models on a new dataset . the proposed model is based on linguistic and semantic information from Chinese conversational speech .

Copied to clipboard

Xinyi Zeng, Xue Yang, Jingyuan Zhang, Huanqian Yan, Xiang Chen, Kaiwen Wei, Hankun Kang, Yu Tian

Sequence Parallelism: Long Sequence Training from System Perspective (2023.acl-long)

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.

Copied to clipboard

Shenggui Li, Fuzhao Xue, Chaitanya Baranwal, Yongbin Li, Yang You

On the Hallucination in Simultaneous Machine Translation (2024.acl-short)

Challenge:	Existing work on memory-efficient parallelisms to reduce time and space complexity focuses on reducing time and complexity from system perspective.
Approach:	They propose a memory-efficient parallelism to reduce time and space complexity . they split input sequence into multiple chunks and feed each chunk into GPU .
Outcome:	The proposed approach is compatible with most existing parallelisms and makes 4D parallelismal possible.

Copied to clipboard

Meizhi Zhong, Kehai Chen, Zhengshan Xue, Lemao Liu, Mingming Yang, Min Zhang

Machine Translation

Challenge:	Currently, there are no studies which systematically analyze hallucination in SiMT.
Approach:	They conduct a comprehensive analysis of hallucination in simultaneous machine translation (SiMT) they find that halluciation is extremely severe, especially as latency increases .
Outcome:	The results show that it is possible to alleviate hallucination by decreasing the over usage of target-side information for SiMT.

LASA: Language-Agnostic Semantic Alignment at the Semantic Bottleneck for LLM Safety (2026.acl-long)

Copied to clipboard

Junxiao Yang, Haoran Liu, Jinzhe Tu, Jiale Cheng, Zhexin Zhang, Shiyao Cui, Jiaqi Weng, Jialing Tao, Hui Xue, Hongning Wang, Han Qiu, Minlie Huang

Challenge:	Large language models (LLMs) have demonstrated better safety performance in high-resource languages than in low-resourced languages.
Approach:	They propose language-agnostic semantic alignment (LASA) which anchors safety alignment directly in semantic bottlenecks.
Outcome:	The proposed approach significantly improves safety across all languages: average attack success rate drops from 24.7% to 2.8% on LLaMA-3.1-8B-Instruct and remains within 3–4% across Qwen2.5 and Qwend3 Instruct models (7B–32B).

Unifying Discrete and Continuous Representations for Unsupervised Paraphrase Generation (2023.emnlp-main)

Copied to clipboard

Mingfeng Xue, Dayiheng Liu, Wenqiang Lei, Jie Fu, Jian Lan, Mei Li, Baosong Yang, Jun Xie, Yidan Zhang, Dezhong Peng, Jiancheng Lv

Text Generation

Challenge:	Existing unsupervised paraphrase generation methods require large-scale, manually annotated paraphrase datasets, which are labor-intensive to build.
Approach:	They propose a self-supervised pseudo-data construction method that generates diverse pseudo-paraphrases in distinct surface structures for a given sentence.
Outcome:	The proposed method generates diverse pseudo-paraphrases in distinct surface structures for a given sentence.

FlashAudio: Rectified Flow for Fast and High-Fidelity Text-to-Audio Generation (2025.acl-long)

Copied to clipboard

Huadai Liu, Jialei Wang, Rongjie Huang, Yang Liu, Heng Lu, Zhou Zhao, Wei Xue

MemWeaver: Weaving Hybrid Memories for Traceable Long-Horizon Agentic Reasoning (2026.findings-acl)

Challenge:	Recent advances in latent diffusion models (LDMs) have markedly enhanced text-to-audio generation, yet their iterative sampling processes impose substantial computational demands, limiting practical deployment.
Approach:	They propose to learn straight flow for fast simulation by using flashAudio with rectified flows and immiscible flow to minimize the total distance of data-noise pairs in a batch vias assignment.
Outcome:	The proposed method can learn straight flow for fast simulations and reduce noise distribution.

Copied to clipboard

Juexiang Ye, Xue Li, Yang Xinyu, Chengkai Huang, Lanshun Nie, Lina Yao, Dechen Zhan

Information Retrieval

Challenge:	Existing methods rely on unstructured retrieval or coarse abstractions, which lead to temporal conflicts, brittle reasoning, and limited traceability.
Approach:	They propose a unified memory framework that consolidates long-term agent experiences into three interconnected components that combine structured knowledge and evidence to construct compact yet information-dense contexts for reasoning.
Outcome:	The proposed framework significantly improves multi-hop and temporal reasoning accuracy while reducing input context length by over 95% compared to long-context baselines.

Seeing but Not Thinking: Routing Distraction in Multimodal Mixture-of-Experts (2026.acl-long)

Copied to clipboard

Haolei Xu, Haiwen Hong, Hongxing Li, Rui Zhou, Yang Zhang, Longtao Huang, Hui Xue, Yongliang Shen, Weiming Lu, Yueting Zhuang

START: Self-taught Reasoner with Tools (2025.emnlp-main)

Challenge:	Existing multimodal Mixture-of-Experts models accurately perceive image content yet fail in subsequent reasoning . Seeing but not thinking phenomenon is a puzzling phenomenon .
Approach:	They propose a routing-guided intervention method that enhances domain expert activation.
Outcome:	The proposed method achieves consistent improvements on visual reasoning tasks.

Copied to clipboard

Chengpeng Li, Mingfeng Xue, Zhenru Zhang, Jiaxi Yang, Beichen Zhang, Bowen Yu, Binyuan Hui, Junyang Lin, Xiang Wang, Dayiheng Liu

Rethinking Token Reduction for State Space Models (2024.emnlp-main)

Challenge:	Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in complex reasoning through long chain-of-thought, yet they struggle with precise computations and algorithmic operations.
Approach:	They propose a training-free approach that activates LRMs’ latent tool-use capabilities through artificial hints and a framework that enables models to learn effective tool utilization through diverse hint patterns and rejection-based data synthesis.
Outcome:	Experiments show that START significantly improves state-of-the-art LRMs across challenging benchmarks, including competition-level mathematics (AMC23: 95.0%, AIME24: 75.6%) and graduate-level science questions (GPQA: 64.6%).

Copied to clipboard

Zheng Zhan, Yushu Wu, Zhenglun Kong, Changdi Yang, Yifan Gong, Xuan Shen, Xue Lin, Pu Zhao, Yanzhi Wang

Challenge:	Existing methods for token reduction for SSMs lead to performance drops . a recent study shows that Mamba-2 improves the accuracy of the model by 5.7% to 13.1% .
Approach:	They propose a token reduction method that integrates token importance and similarity into SSMs and takes advantage of pruning and merging.
Outcome:	The proposed method improves accuracy by 5.7% to 13.1% on six benchmarks with Mamba-2 compared to existing methods while reducing computational demands and memory requirements.

GCPG: A General Framework for Controllable Paraphrase Generation (2022.findings-acl)

Copied to clipboard

Kexin Yang, Dayiheng Liu, Wenqiang Lei, Baosong Yang, Haibo Zhang, Xue Zhao, Wenqing Yao, Boxing Chen

Text Generation

Challenge:	Existing studies highlight a special condition under two indispensable aspects of controllable paraphrase generation (CPG) individually, lacking a unified circumstance to explore and analyze their effectiveness.
Approach:	They propose a general controllable paraphrase generation framework that integrates lexical and syntactical conditions into a text sequence and uniformly processes them in an encoder-decoder paradigm.
Outcome:	The proposed framework can combine lexical and syntactical conditions and improve paraphrase generation.

Transformer-GCRF: Recovering Chinese Dropped Pronouns with General Conditional Random Fields (2020.findings-emnlp)

Copied to clipboard

Jingxuan Yang, Kerui Xu, Jun Xu, Si Li, Sheng Gao, Jun Guo, Ji-Rong Wen, Nianwen Xue

Sherry: Hardware-Efficient 1.25-Bit Ternary Quantization via Fine-grained Sparsification (2026.acl-long)

Challenge:	Existing approaches to recover dropped pronouns ignore the dependencies between pronounes in neighboring utterances.
Approach:	They propose a framework that combines Transformer network and General Conditional Random Fields to model the dependencies between pronouns in neighboring utterances.
Outcome:	The proposed framework outperforms state-of-the-art models on three Chinese conversation datasets showing that it captures the dependencies between pronouns in neighboring utterances.

Copied to clipboard

Hong Huang, Decheng Wu, Qiangqiang Hu, Guanghua Yu, Jinhai Yang, Jianchen Zhu, Xue Liu, Dapeng Wu

Green & Sustainable NLP

Challenge:	ternary quantization is a powerful solution for resource-constrained edge devices . current implementations suffer from a fundamental misalignment with commodity hardware .
Approach:	They propose a hardware-efficient ternary quantization framework that packs weights into five bits to restore power-of-two alignment.
Outcome:	The proposed framework reduces weights to -1, 0, +1 while preserving power-of-two alignment.

Generating Authentic Adversarial Examples beyond Meaning-preserving with Doubly Round-trip Translation (2022.naacl-main)

Copied to clipboard

Siyu Lai, Zhen Yang, Fandong Meng, Xue Zhang, Yufeng Chen, Jinan Xu, Jie Zhou

Robustness in NLP

Challenge:	Existing approaches to generate adversarial examples for NMT use the meaning-preserving restriction.
Approach:	They propose a new definition for adversarial examples based on the Doubly Round-Trip Translation (DRTT) they introduce masked language models to construct bilingual adversarials based upon DRTT .
Outcome:	The proposed approach significantly improves the robustness of the NMT model on clean and noisy test sets.

EoT: Evolution of Thoughts for Complex Reasoning Tasks (2025.findings-emnlp)

Copied to clipboard

Qin Hua, Jiaqi Sun, Shiyou Qian, Dingyu Yang, Jian Cao, Guangtao Xue

Boosting Policy and Process Reward Models with Monte Carlo Tree Search in Open-Domain QA (2025.findings-acl)

Challenge:	Existing studies focus on ensuring behavior fidelity, factuality or reliability in generated reasoning processes, but they neglect the simultaneous optimization of all three aspects for each thought.
Approach:	They propose a thought assessment method that is sensitive to knowledge and LLM behaviors . they use three scorers to evaluate each thought by considering domain context, semantic alignment, and behavior impact.
Outcome:	The proposed framework outperforms advanced approaches in knowledge-based complex tasks.

Copied to clipboard

Chi-Min Chan, Chunpu Xu, Junqi Zhu, Jiaming Ji, Donghai Hong, Pengcheng Wen, Chunyang Jiang, Zhen Ye, Yaodong Yang, Wei Xue, Sirui Han, Yike Guo

Question Answering

Challenge:	Experimental results show that our approach can effectively improve the performance of both the policy model and the reward model.
Approach:	They propose to use Monte Carlo Tree Search for both policy model improvement and reward model improvement to bridge it to more subtle open-domain question answering.
Outcome:	The proposed approach surpasses existing methods for annotation and training data with fewer data points and achieves better performance in test-time scaling strategies.

DPN-LE: Dual Personality Neuron Localization and Editing for Large Language Models (2026.findings-acl)

Copied to clipboard

Lifan Zheng, Xue Yang, Jiawei Chen, Chenyan WU, Jingyuan Zhang, Fanheng Kong, Xinyi Zeng, Xiang Chen, Yu Tian

SparkRA: A Retrieval-Augmented Knowledge Service System Based on Spark Large Language Model (2024.emnlp-demo)

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.

Copied to clipboard

Dayong Wu, Jiaqi Li, Baoxin Wang, Honghong Zhao, Siyuan Xue, Yanjie Yang, Zhijun Chang, Rui Zhang, Li Qian, Bo Wang, Shijin Wang, Zhixiong Zhang, Guoping Hu

Multimodal Emotion Recognition in Conversations: A Survey of Methods, Trends, Challenges and Prospects (2025.findings-emnlp)

Challenge:	Large language models (LLMs) have shown remarkable achievements across various language tasks.
Approach:	They propose a scientific literature LLM and a knowledge service system based on it . they collect scientific literature and then pre-train it using autoregressive training .
Outcome:	The proposed system provides literature investigation, paper reading, and academic writing functions.

Copied to clipboard

ChengYan Wu, Yiqiang Cai, Yang Liu, Pengxu Zhu, Yun Xue, Ziwei Gong, Julia Hirschberg, Bolei Ma

Sentiment Analysis

Challenge:	Multimodal Emotion Recognition in Conversations (MERC) is a new way to enhance human-computer interaction.
Approach:	This survey offers a systematic overview of Multimodal Emotion Recognition in Conversations . it examines motivations, core tasks, representative methods, and evaluation strategies .
Outcome:	The survey examines the effectiveness of MERC and its evaluation strategies.

SIV-Bench: A Video Benchmark for Social Interaction Understanding and Reasoning (2026.findings-acl)

Copied to clipboard

Fanqi Kong, Weiqin Zu, Xinyu Chen, Yaodong Yang, Song-Chun Zhu, Xue Feng

Recovering dropped pronouns in Chinese conversations via modeling their referents (N19-1)

Challenge:	Existing benchmarks for Multimodal Large Language Models (MLLMs) have been lacking due to the rich nature of social interaction.
Approach:	They propose a video benchmark to evaluate MLLMs' capabilities across social scene understanding, social state reasoning, and social dynamics prediction.
Outcome:	The proposed benchmarks evaluate MLLMs' capabilities across social scene understanding, social state reasoning, and social dynamics prediction tasks.

Copied to clipboard

Jingxuan Yang, Jianzhuo Tong, Si Li, Sheng Gao, Jun Guo, Nianwen Xue

Dialogue Systems & Conversational Agents

Challenge:	Pronouns are often dropped in conversational genres as their referents can be easily understood from context.
Approach:	They propose an end-to-end neural network model to recover dropped pronouns in conversational data.
Outcome:	The proposed model improves on three different conversational genres.

Rust-doctor: Enhanced Feature for Rust Ownership and Lifetime Repair with Balanced Training Data Generation (2025.findings-emnlp)

Copied to clipboard

Wenzhang Yang, Xiaoning Ren, Cuifeng Gao, Yinxing Xue

Tailor: A Soft-Prompt-Based Approach to Attribute-Based Controlled Text Generation (2023.acl-long)

Challenge:	Rust has gained popularity due to its safety features during compilation.
Approach:	They propose to use a training dataset to generate a balanced training dataset and leverage the Mid-level Intermediate Representation (MIR) as enhanced features for Rust’s unique compilation error repair.
Outcome:	The proposed model outperforms state-of-the-art models in terms of Pass@K and Acc@K.

Copied to clipboard

Kexin Yang, Dayiheng Liu, Wenqiang Lei, Baosong Yang, Mingfeng Xue, Boxing Chen, Jun Xie

CONSTRUCTA: Automating Commercial Construction Schedules in Fabrication Facilities with Large Language Models (2025.naacl-industry)

Challenge:	Existing work focuses on generating sentences satisfying pre-specified attributes such as topic and sentiment, yet suffers from increases in storage and inference time.
Approach:	They propose a method that uses a pre-trained continuous vector to generate a fixed pre-trainable language model to satisfy a specified attribute.
Outcome:	The proposed model can achieve improvements on eleven attribute-specific generation tasks with 0.08% extra training parameters.

Copied to clipboard

Yifan Zhang, Xue Yang

We Need to Talk About Reproducibility in NLP Model Comparison (2023.emnlp-main)

Challenge:	Existing methods for automating construction scheduling are limited due to their domain knowledge and complexity.
Approach:	They propose a framework leveraging LLMs to optimize construction schedules in commercial construction.
Outcome:	The proposed framework improves missing value prediction, dependency analysis and planning performance compared to baseline methods.

Copied to clipboard

Yan Xue, Xuefei Cao, Xingli Yang, Yu Wang, Ruibo Wang, Jihong Li

Red Teaming Large Reasoning Models (2026.acl-long)

Challenge:	Existing studies show that standard splits produce low reproducible and unreliable conclusions . reproducibility of empirical experimental conclusions is a problem in NLP domain .
Approach:	They propose to transform the reproducibility of a model comparison into a probabilistic function . they propose to use a regularized corpus splitting strategy to estimate the model's performance .
Outcome:	The proposed estimator achieves a high SNR and significantly increases reproducibility.

Copied to clipboard

Jiawei Chen, Yang Yang, Chao Yu, Yu Tian, Zhi Cao, Xue Yang, Linghao Li, Hang Su, Zhaoxia Yin

AscendKernelGen: LLM-Driven Kernel Generation for NPUs (2026.findings-acl)

Challenge:	Large Reasoning Models (LRMs) have emerged as a powerful advancement in multi-step reasoning tasks, but they introduce safety and reliability risks, such as CoT-hijacking and prompt-induced inefficiencies.
Approach:	They propose a unified benchmark to assess the trustworthiness of Large Reasoning Models.
Outcome:	The proposed benchmark evaluates truthfulness, safety and efficiency on 26 models.

Copied to clipboard

Xinzi Cao, Jianyang Zhai, Pengfei Li, Zhiheng Hu, Cen Yan, null Mubingxu, Guanghuan Fang, Bin She, Jiayu Li, Yihan Su, Dongyang Tao, Feidiao Yang, Chang-Dong Wang, Yutong Lu, Weicheng Xue, Bin Zhou, Yonghong Tian

Challenge:	Neural Processing Units (NPUs) are critical for AI infrastructure, but their development remains a bottleneck due to vendor-specific Domain-Specific Languages (DSLs).
Approach:	They propose a framework for NPU kernel development that bridges the gap in hardware-specific coding . compiler success on complex Level-2 kernels improves from 0% to 95.5%, they say .
Outcome:	The proposed framework bridges the gap in hardware-specific coding, showing a near-zero success rate on complex kernels.

E-Verify: A Paradigm Shift to Scalable Embedding-based Factuality Verification (2025.findings-emnlp)

Copied to clipboard

Zeyang Liu, Jingfeng Xue, Xiuqi Yang, Wenbiao Du, Jiarun Fu, Junbao Chen, Wenjie Guo, Yong Wang

Generalization or Memorization: Data Contamination and Trustworthy Evaluation for Large Language Models (2024.findings-acl)

Challenge:	Existing factuality verification methods follow a Decompose-Then-Verify paradigm, which improves granularity but suffers from poor scalability and efficiency.
Approach:	They propose a Decompose-Embed-Interact paradigm that shifts factuality verification from costly text-level reasoning to efficient alignment in embedding space.
Outcome:	The proposed paradigm shifts factuality verification from costly text-level reasoning to efficient alignment in embedding space .

Copied to clipboard

Yihong Dong, Xue Jiang, Huanyu Liu, Zhi Jin, Bin Gu, Mengfei Yang, Ge Li

V-Oracle: Making Progressive Reasoning in Deciphering Oracle Bones for You and Me (2025.acl-long)

Challenge:	Considering the vast size and wide-ranging sources of LLMs’ training data, it could explicitly or implicitly include test data.
Approach:	They propose a Contamination Detection via output Distribution (CDD) which detects data contamination only by identifying the peakedness of LLM's output distribution.
Outcome:	The proposed method improves performance by 21.8%-30.2% on humanEval and TED: trustworthy evaluation via output distribution.

Copied to clipboard

Runqi Qiao, Qiuna Tan, Guanting Dong, MinhuiWu MinhuiWu, Jiapeng Wang, YiFan Zhang, Zhuoma GongQue, Chong Sun, Yida Xu, Yadong Xue, Ye Tian, Zhimin Bao, Lan Yang, Chen Li, Honggang Zhang

Verifiable LLM-Generated Text Detection via Projected Semantic-Structural Distributions (2026.acl-long)

Challenge:	Deciphering oracle bone scripts using AI technology is not an overnight task due to the evolution of written language over millennia.
Approach:	They propose a framework that utilizes Large Multi-modal Models (LMMs) for interpreting Oracle Bone Script (OBS).
Outcome:	The proposed framework provides quantitative analyses and superior deciphering capability.

Copied to clipboard

Ruochong Xiong, Qien Li, Wangwang Lian, Yulong Wan, Hanlin Xue, Zhouxing Tan, Han Yang, Fengyu Lu, Junfei Liu

Text Generation

Challenge:	Existing methods for detecting LLM-Generated text suffer from distribution misalignment and limited interpretability.
Approach:	They propose a statistical framework utilizing supervised subspace learning to extract compact features and construct conditional semantic distributions based on syntactic structures.
Outcome:	The proposed framework is superior in cross-domain, cross-model, and adversarial scenarios.

Dynamic Voting for Efficient Reasoning in Large Language Models (2023.findings-emnlp)

Copied to clipboard

Mingfeng Xue, Dayiheng Liu, Wenqiang Lei, Xingzhang Ren, Baosong Yang, Jun Xie, Yidan Zhang, Dezhong Peng, Jiancheng Lv

Communication-Efficient and Tensorized Federated Fine-Tuning of Large Language Models (2025.findings-acl)

Challenge:	Multi-path voting methods generate multiple reasoning paths for each problem, causing factual errors and illusion generation.
Approach:	They propose a multi-path voting technique that effectively reduces the number of reasoning paths during multi-path voting while preserving accuracies.
Outcome:	The proposed method outperforms Self-consistency using 24.7% of the number of paths on the LetterConcat task.

Copied to clipboard

Sajjad Ghiasvand, Yifan Yang, Zhiyu Xue, Mahnoosh Alizadeh, Zheng Zhang, Ramtin Pedarsani

MMSciCode: Real-world Evaluation of Multilingual Multi-Discipline Scientific Research Coding (2026.acl-long)

Challenge:	Large Language Models (LLMs) excel in translation and summarization due to the capabilities of transformer architectures.
Approach:	They propose to integrate tensorized adapters into model encoder/decoder blocks to improve model adaptability against data heterogeneity.
Outcome:	Experiments on large-scale cross-device FL and large-silo FL show that the proposed methods perform on par or even better than existing federated PEFT approaches while reducing communication cost.

Copied to clipboard

Xue Xia, Zheyuan Yang, Arman Cohan, Yilun Zhao

Programming Languages in NLP

Challenge:	MMSciCode is a benchmark for evaluating foundation models in scientific code generation.
Approach:	They propose a multilingual, multi-discipline benchmark for evaluating foundation models in scientific code generation that integrates domain-specific knowledge with algorithmic reasoning.
Outcome:	The new benchmark is annotated by domain experts and features rigorous quality controls to ensure dataset integrity and authenticity.

QuantumQA: Enhancing Scientific Reasoning via Physics-Consistent Dataset and Verification-Aware Reinforcement Learning (2026.acl-long)

Copied to clipboard

Songxin Qu, Tai-Ping Sun, Yun-Jie Wang, Huan-Yu Liu, Cheng Xue, Xiao-Fan Xu, Han Fang, Yang Yang, Yu-Chun Wu, Guo-Ping Guo, Zhao-Yun Chen

Enhancing LLM-Based Social Bot via an Adversarial Learning Framework (2025.emnlp-main)

Challenge:	Large language models lack reliability in scientific domains that require strict adherence to physical constraints.
Approach:	They propose a large-scale dataset constructed via a task-adaptive strategy and a hybrid verification protocol that combines deterministic solvers with semantic auditing to guarantee scientific rigor.
Outcome:	The proposed model outperforms baselines and general-purpose preference models and is competitive with proprietary models.

Copied to clipboard

Fanqi Kong, Xiaoyuan Zhang, Xinyu Chen, Yaodong Yang, Song-Chun Zhu, Xue Feng

Robustness in NLP

Challenge:	Social media platforms provide an ideal testbed for large language models that exhibit human-like behavior.
Approach:	They propose an LLM-based social **Bot that enhances human-like generative capabilities through an adversarial learning framework.
Outcome:	The proposed framework generates human-like content aligned with diverse user profiles . it exhibits strong social responsiveness, more accurately modeling opinion dynamics .

OSCR-Attack: One-Shot Character Level Attacks through Self-Optimizing Continuous Relaxation (2026.findings-acl)

Copied to clipboard

Lingyi Kong, Zhuo Liu, Zhanghao Hu, Qilong Qiu, Yutao Yang, Jingjing Xue, Zheng Wang, Lin Gui, Feiping Nie

Robustness in NLP

Challenge:	Character-level adversarial attacks preserve semantics but are costly and inefficient . generative LLMs are gaining popularity due to their uncertainty and vulnerability to textual adversarials .
Approach:	They propose an end-to-end framework that transforms discrete choices into continuous representations and a conflict resolution strategy that maps them back into discrete insertion operations.
Outcome:	The proposed framework improves ASR by 21.45% points and accelerates the attack by 3.66 times compared to baselines.

DiffER: Diffusion Entity-Relation Modeling for Reversal Curse in Diffusion Large Language Models (2026.findings-acl)

Copied to clipboard

Shaokai He, Kaiwen Wei, Xinyi Zeng, Xiang Chen, Xue Yang, Zhenyang Li, Jiang Zhong, Yu Tian

Can LLMs Generate High-Quality Test Cases for Algorithm Problems? TestCase-Eval: A Systematic Evaluation of Fault Coverage and Exposure (2025.acl-short)

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 .

Copied to clipboard

Zheyuan Yang, Zexi Kuang, Xue Xia, Yilun Zhao

Challenge:	TestCase-Eval focuses on Fault Coverage and Fault Exposure tasks . authors provide insights into their strengths and limitations in generating effective test cases . correctness and robustness of algorithmic solutions hinge on quality of test suites .
Approach:	They introduce TestCase-Eval, a benchmark for systematic evaluation of LLMs in test-case generation.
Outcome:	The new benchmark measures the performance of LLMs in test-case generation.

Safe-FedLLM: Delving into the Safety of Federated Large Language Models (2026.acl-long)

Copied to clipboard

Mingxiang Tao, Yu Tian, Wenxuan Tu, Yue Yang, Xue Yang, Xiangyan Tang

Text Classification

Challenge:	Existing work on federated learning for large language models (FL) addresses privacy and data-silo issues in the training of large language model training.
Approach:	They propose a probe-based defense framework for FedLLM that constructs defenses across three levels: Step-Level, Client-Level and Shadow-Level.
Outcome:	The proposed framework improves FedLLM's robustness against malicious clients while maintaining competitive performance on benign data.

Dynamic Evil Score-Guided Decoding: An Efficient Decoding Framework For Red-Team Model (2025.findings-acl)

Copied to clipboard

Cong Gao, Bo Zhang, Linkang Yang, Minghao Hu, Zhunchen Luo, Xiaoying Bai, Guotong Geng, Jun Zhang, Yunhua Xue

Unified Hallucination Detection for Multimodal Large Language Models (2024.acl-long)

Challenge:	Existing red-teaming methods require expensive fine-tuning, especially for large LLMs.
Approach:	They propose a red-teaming method that uses an ‘evil score’ to evaluate the potential of tokens to contribute to harmful outputs during decoding.
Outcome:	The proposed method achieves an ASR of 92.83% on the Llama-3.2-3B-Instruct model, compared to 83.48% with adversarial fine-tuning while using less computational resources.

Copied to clipboard

Xiang Chen, Chenxi Wang, Yida Xue, Ningyu Zhang, Xiaoyan Yang, Qiang Li, Yue Shen, Lei Liang, Jinjie Gu, Huajun Chen

The Emperor’s New Reasoning: Format Imitation Overshadows Genuine Mathematical Understanding in SFT (2025.emnlp-main)

Challenge:	despite significant strides in multimodal tasks, MLLMs are plagued by the critical issue of hallucination.
Approach:	They propose a meta-evaluation benchmark to facilitate evaluation of advancements in hallucination detection methods.
Outcome:	The proposed framework validates hallucinations robustly and provides strategic insights . MHaluBench is a meta-evaluation benchmark designed to facilitate evaluation .

Copied to clipboard

Linyao Yang, Jian-Tao Huang, Yafei Lu, Zhenhui Jessie Li, Guirong Xue