Li Zhou

Li Zhou, Zikun Zhou, Kaige Mao et al. · amazon-science

Tracking by natural language specification aims to locate the referred target in a sequence based on the natural language description. Existing algorithms solve this issue in two steps, visual grounding and tracking, and accordingly deploy the separated grounding model and tracking model to implement these two steps, respectively. Such a separated framework overlooks the link between visual grounding and tracking, which is that the natural language descriptions provide global semantic cues for localizing the target for both two steps. Besides, the separated framework can hardly be trained end-to-end. To handle these issues, we propose a joint visual grounding and tracking framework, which reformulates grounding and tracking as a unified task: localizing the referred target based on the given visual-language references. Specifically, we propose a multi-source relation modeling module to effectively build the relation between the visual-language references and the test image. In addition, we design a temporal modeling module to provide a temporal clue with the guidance of the global semantic information for our model, which effectively improves the adaptability to the appearance variations of the target. Extensive experimental results on TNL2K, LaSOT, OTB99, and RefCOCOg demonstrate that our method performs favorably against state-of-the-art algorithms for both tracking and grounding. Code is available at https://github.com/lizhou-cs/JointNLT.

Yining Hua, Hang Jiang, Shixu Lin et al. · amazon-science, harvard

Understanding public discourse on emergency use of unproven therapeutics is crucial for monitoring safe use and combating misinformation. We developed a natural language processing-based pipeline to comprehend public perceptions of and stances on coronavirus disease 2019 (COVID-19)-related drugs on Twitter over time. This retrospective study included 609,189 US-based tweets from January 29, 2020, to November 30, 2021, about four drugs that garnered significant public attention during the COVID-19 pandemic: (1) Hydroxychloroquine and Ivermectin, therapies with anecdotal evidence; and (2) Molnupiravir and Remdesivir, FDA-approved treatments for eligible patients. Time-trend analysis was employed to understand popularity trends and related events. Content and demographic analyses were conducted to explore potential rationales behind people's stances on each drug. Time-trend analysis indicated that Hydroxychloroquine and Ivermectin were discussed more than Molnupiravir and Remdesivir, particularly during COVID-19 surges. Hydroxychloroquine and Ivermectin discussions were highly politicized, related to conspiracy theories, hearsay, and celebrity influences. The distribution of stances between the two major US political parties was significantly different (P < .001); Republicans were more likely to support Hydroxychloroquine (55%) and Ivermectin (30%) than Democrats. People with healthcare backgrounds tended to oppose Hydroxychloroquine (7%) more than the general population, while the general population was more likely to support Ivermectin (14%). Our study found that social media users have varying perceptions and stances on off-label versus FDA-authorized drug use at different stages of COVID-19. This indicates that health systems, regulatory agencies, and policymakers should design tailored strategies to monitor and reduce misinformation to promote safe drug use.

Antonia Karamolegkou, Jiaang Li, Li Zhou et al.

Language models may memorize more than just facts, including entire chunks of texts seen during training. Fair use exemptions to copyright laws typically allow for limited use of copyrighted material without permission from the copyright holder, but typically for extraction of information from copyrighted materials, rather than {\em verbatim} reproduction. This work explores the issue of copyright violations and large language models through the lens of verbatim memorization, focusing on possible redistribution of copyrighted text. We present experiments with a range of language models over a collection of popular books and coding problems, providing a conservative characterization of the extent to which language models can redistribute these materials. Overall, this research highlights the need for further examination and the potential impact on future developments in natural language processing to ensure adherence to copyright regulations. Code is at \url{https://github.com/coastalcph/CopyrightLLMs}.

Li Zhou, Marc Dacier, Charalambos Konstantinou

The Software Bill of Materials (SBOM) is a critical tool for securing the software supply chain (SSC), but its practical utility is undermined by inaccuracies in both its generation and its application in vulnerability scanning. This paper presents a large-scale empirical study on 2,414 open-source repositories to address these issues from a practical standpoint. First, we demonstrate that using lock files with strong package managers enables the generation of accurate and consistent SBOMs, establishing a reliable foundation for security analysis. Using this high-fidelity foundation, however, we expose a more fundamental flaw in practice: downstream vulnerability scanners produce a staggering 92.0\% false positive rate in our case study. We pinpoint the primary cause as the flagging of vulnerabilities within unreachable code. We then demonstrate that function call analysis can effectively prune 61.9\% of these false alarms. Our work validates a practical, two-stage approach for SSC security: first, generate an accurate SBOM using lock files and strong package managers, and second, enrich it with function call analysis to produce actionable, low-noise vulnerability reports that alleviate developers' alert fatigue.

Jianqiang Xia, DianXi Shi, Ke Song et al.

Most existing RGB-T tracking networks extract modality features in a separate manner, which lacks interaction and mutual guidance between modalities. This limits the network's ability to adapt to the diverse dual-modality appearances of targets and the dynamic relationships between the modalities. Additionally, the three-stage fusion tracking paradigm followed by these networks significantly restricts the tracking speed. To overcome these problems, we propose a unified single-stage Transformer RGB-T tracking network, namely USTrack, which unifies the above three stages into a single ViT (Vision Transformer) backbone with a dual embedding layer through self-attention mechanism. With this structure, the network can extract fusion features of the template and search region under the mutual interaction of modalities. Simultaneously, relation modeling is performed between these features, efficiently obtaining the search region fusion features with better target-background discriminability for prediction. Furthermore, we introduce a novel feature selection mechanism based on modality reliability to mitigate the influence of invalid modalities for prediction, further improving the tracking performance. Extensive experiments on three popular RGB-T tracking benchmarks demonstrate that our method achieves new state-of-the-art performance while maintaining the fastest inference speed 84.2FPS. In particular, MPR/MSR on the short-term and long-term subsets of VTUAV dataset increased by 11.1$\%$/11.7$\%$ and 11.3$\%$/9.7$\%$.

Yong Cao, Li Zhou, Seolhwa Lee et al.

The recent release of ChatGPT has garnered widespread recognition for its exceptional ability to generate human-like responses in dialogue. Given its usage by users from various nations and its training on a vast multilingual corpus that incorporates diverse cultural and societal norms, it is crucial to evaluate its effectiveness in cultural adaptation. In this paper, we investigate the underlying cultural background of ChatGPT by analyzing its responses to questions designed to quantify human cultural differences. Our findings suggest that, when prompted with American context, ChatGPT exhibits a strong alignment with American culture, but it adapts less effectively to other cultural contexts. Furthermore, by using different prompts to probe the model, we show that English prompts reduce the variance in model responses, flattening out cultural differences and biasing them towards American culture. This study provides valuable insights into the cultural implications of ChatGPT and highlights the necessity of greater diversity and cultural awareness in language technologies.

Yining Hua, Jiageng Wu, Shixu Lin et al.

Objective: Social media-based public health research is crucial for epidemic surveillance, but most studies identify relevant corpora with keyword-matching. This study develops a system to streamline the process of curating colloquial medical dictionaries. We demonstrate the pipeline by curating a UMLS-colloquial symptom dictionary from COVID-19-related tweets as proof of concept. Methods: COVID-19-related tweets from February 1, 2020, to April 30, 2022 were used. The pipeline includes three modules: a named entity recognition module to detect symptoms in tweets; an entity normalization module to aggregate detected entities; and a mapping module that iteratively maps entities to Unified Medical Language System concepts. A random 500 entity sample were drawn from the final dictionary for accuracy validation. Additionally, we conducted a symptom frequency distribution analysis to compare our dictionary to a pre-defined lexicon from previous research. Results: We identified 498,480 unique symptom entity expressions from the tweets. Pre-processing reduces the number to 18,226. The final dictionary contains 38,175 unique expressions of symptoms that can be mapped to 966 UMLS concepts (accuracy = 95%). Symptom distribution analysis found that our dictionary detects more symptoms and is effective at identifying psychiatric disorders like anxiety and depression, often missed by pre-defined lexicons. Conclusions: This study advances public health research by implementing a novel, systematic pipeline for curating symptom lexicons from social media data. The final lexicon's high accuracy, validated by medical professionals, underscores the potential of this methodology to reliably interpret and categorize vast amounts of unstructured social media data into actionable medical insights across diverse linguistic and regional landscapes.

Heather Lent, Kushal Tatariya, Raj Dabre et al.

Creoles represent an under-explored and marginalized group of languages, with few available resources for NLP research.While the genealogical ties between Creoles and a number of highly-resourced languages imply a significant potential for transfer learning, this potential is hampered due to this lack of annotated data. In this work we present CreoleVal, a collection of benchmark datasets spanning 8 different NLP tasks, covering up to 28 Creole languages; it is an aggregate of novel development datasets for reading comprehension, relation classification, and machine translation for Creoles, in addition to a practical gateway to a handful of preexisting benchmarks. For each benchmark, we conduct baseline experiments in a zero-shot setting in order to further ascertain the capabilities and limitations of transfer learning for Creoles. Ultimately, we see CreoleVal as an opportunity to empower research on Creoles in NLP and computational linguistics, and in general, a step towards more equitable language technology around the globe.

Li Zhou, Laura Cabello, Yong Cao et al.

Detecting offensive language is a challenging task. Generalizing across different cultures and languages becomes even more challenging: besides lexical, syntactic and semantic differences, pragmatic aspects such as cultural norms and sensitivities, which are particularly relevant in this context, vary greatly. In this paper, we target Chinese offensive language detection and aim to investigate the impact of transfer learning using offensive language detection data from different cultural backgrounds, specifically Korean and English. We find that culture-specific biases in what is considered offensive negatively impact the transferability of language models (LMs) and that LMs trained on diverse cultural data are sensitive to different features in Chinese offensive language detection. In a few-shot learning scenario, however, our study shows promising prospects for non-English offensive language detection with limited resources. Our findings highlight the importance of cross-cultural transfer learning in improving offensive language detection and promoting inclusive digital spaces.

Li Zhou, Antonia Karamolegkou, Wenyu Chen et al.

The increasing ubiquity of language technology necessitates a shift towards considering cultural diversity in the machine learning realm, particularly for subjective tasks that rely heavily on cultural nuances, such as Offensive Language Detection (OLD). Current understanding underscores that these tasks are substantially influenced by cultural values, however, a notable gap exists in determining if cultural features can accurately predict the success of cross-cultural transfer learning for such subjective tasks. Addressing this, our study delves into the intersection of cultural features and transfer learning effectiveness. The findings reveal that cultural value surveys indeed possess a predictive power for cross-cultural transfer learning success in OLD tasks and that it can be further improved using offensive word distance. Based on these results, we advocate for the integration of cultural information into datasets. Additionally, we recommend leveraging data sources rich in cultural information, such as surveys, to enhance cultural adaptability. Our research signifies a step forward in the quest for more inclusive, culturally sensitive language technologies.

Yingying Cheng, Jinquan Shi, Li Zhou et al.

Quantization-aware training (QAT) with low-bit floating-point formats enables efficient LLM deployment, yet introduces subtle failure modes invisible to standard training metrics. We present a systematic study of HiF8 W8A8 QAT for OpenPangu-Embedded-1B through the lens of Delayed Tensor Scaling (DTS). Across eight controlled experiments, we identify and disentangle two orthogonal failure modes: (i)amax saturation, where delayed scale estimates silently corrupt knowledge-sensitive representations via forward-pass clipping, and (ii)catastrophic forgetting, where an aggressive learning rate overwrites pretrained commonsense knowledge independently of quantization. Neither is detectable from training loss alone. We address amax saturation with a conservative max-algorithm DTS strategy over a 64-step history window, and mitigate forgetting via a 500-step BF16 warmup followed by QAT at lr=10^{-5}. Both fixes are necessary and sufficient: our final configuration achieves 0.43% MMLU drop, 0.58% HellaSwag drop, and 0.22% ARC-Challenge drop versus a matched BF16 baseline, with a training loss APE of only 0.11% over 10,000 steps.

Yong Cao, Yova Kementchedjhieva, Ruixiang Cui et al.

Building upon the considerable advances in Large Language Models (LLMs), we are now equipped to address more sophisticated tasks demanding a nuanced understanding of cross-cultural contexts. A key example is recipe adaptation, which goes beyond simple translation to include a grasp of ingredients, culinary techniques, and dietary preferences specific to a given culture. We introduce a new task involving the translation and cultural adaptation of recipes between Chinese and English-speaking cuisines. To support this investigation, we present CulturalRecipes, a unique dataset comprised of automatically paired recipes written in Mandarin Chinese and English. This dataset is further enriched with a human-written and curated test set. In this intricate task of cross-cultural recipe adaptation, we evaluate the performance of various methods, including GPT-4 and other LLMs, traditional machine translation, and information retrieval techniques. Our comprehensive analysis includes both automatic and human evaluation metrics. While GPT-4 exhibits impressive abilities in adapting Chinese recipes into English, it still lags behind human expertise when translating English recipes into Chinese. This underscores the multifaceted nature of cultural adaptations. We anticipate that these insights will significantly contribute to future research on culturally-aware language models and their practical application in culturally diverse contexts.

Feng Chen, Liqin Wang, Julie Hong et al.

Objectives: Leveraging artificial intelligence (AI) in conjunction with electronic health records (EHRs) holds transformative potential to improve healthcare. Yet, addressing bias in AI, which risks worsening healthcare disparities, cannot be overlooked. This study reviews methods to detect and mitigate diverse forms of bias in AI models developed using EHR data. Methods: We conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines, analyzing articles from PubMed, Web of Science, and IEEE published between January 1, 2010, and Dec 17, 2023. The review identified key biases, outlined strategies for detecting and mitigating bias throughout the AI model development process, and analyzed metrics for bias assessment. Results: Of the 450 articles retrieved, 20 met our criteria, revealing six major bias types: algorithmic, confounding, implicit, measurement, selection, and temporal. The AI models were primarily developed for predictive tasks in healthcare settings. Four studies concentrated on the detection of implicit and algorithmic biases employing fairness metrics like statistical parity, equal opportunity, and predictive equity. Sixty proposed various strategies for mitigating biases, especially targeting implicit and selection biases. These strategies, evaluated through both performance (e.g., accuracy, AUROC) and fairness metrics, predominantly involved data collection and preprocessing techniques like resampling, reweighting, and transformation. Discussion: This review highlights the varied and evolving nature of strategies to address bias in EHR-based AI models, emphasizing the urgent needs for the establishment of standardized, generalizable, and interpretable methodologies to foster the creation of ethical AI systems that promote fairness and equity in healthcare.

Ailin Huang, Boyong Wu, Bruce Wang et al.

Real-time speech interaction, serving as a fundamental interface for human-machine collaboration, holds immense potential. However, current open-source models face limitations such as high costs in voice data collection, weakness in dynamic control, and limited intelligence. To address these challenges, this paper introduces Step-Audio, the first production-ready open-source solution. Key contributions include: 1) a 130B-parameter unified speech-text multi-modal model that achieves unified understanding and generation, with the Step-Audio-Chat version open-sourced; 2) a generative speech data engine that establishes an affordable voice cloning framework and produces the open-sourced lightweight Step-Audio-TTS-3B model through distillation; 3) an instruction-driven fine control system enabling dynamic adjustments across dialects, emotions, singing, and RAP; 4) an enhanced cognitive architecture augmented with tool calling and role-playing abilities to manage complex tasks effectively. Based on our new StepEval-Audio-360 evaluation benchmark, Step-Audio achieves state-of-the-art performance in human evaluations, especially in terms of instruction following. On open-source benchmarks like LLaMA Question, shows 9.3% average performance improvement, demonstrating our commitment to advancing the development of open-source multi-modal language technologies. Our code and models are available at https://github.com/stepfun-ai/Step-Audio.

Qifan Huang, Minbo Gao, Li Zhou et al.

Most quantum compilers assume programs are reversible unitary circuits. This fits closed-system algorithms, but not open-system simulation, where the natural program objects are quantum channels describing non-unitary dynamics. We present a channel-first compilation framework that treats channels as first-class compilation objects. Our core IR, ChannelIR, represents channels explicitly in Kraus form, a standard channel representation, with Pauli-sum structure, enabling algebraic rewrites before circuit synthesis. We instantiate the framework with LindFront, a frontend that lowers continuous-time Lindbladian generators to short-time channels, and a backend that compiles these channels to executable circuits with structure-aware optimizations. On Lindbladian and channel-simulation benchmarks, the optimized pipeline reduces gate count by up to 99% over an unoptimized channel-first baseline and scales better than circuit-first Stinespring compilation.

Boyong Wu, Chao Yan, Chen Hu et al.

This paper presents Step-Audio 2, an end-to-end multi-modal large language model designed for industry-strength audio understanding and speech conversation. By integrating a latent audio encoder and reasoning-centric reinforcement learning (RL), Step-Audio 2 achieves promising performance in automatic speech recognition (ASR) and audio understanding. To facilitate genuine end-to-end speech conversation, Step-Audio 2 incorporates the generation of discrete audio tokens into language modeling, significantly enhancing its responsiveness to paralinguistic information such as speaking styles and emotions. To effectively leverage the rich textual and acoustic knowledge in real-world data, Step-Audio 2 integrates retrieval-augmented generation (RAG) and is able to call external tools such as web search to mitigate hallucination and audio search to switch timbres. Trained on millions of hours of speech and audio data, Step-Audio 2 delivers intelligence and expressiveness across diverse conversational scenarios. Evaluation results demonstrate that Step-Audio 2 achieves state-of-the-art performance on various audio understanding and conversational benchmarks compared to other open-source and commercial solutions. Please visit https://github.com/stepfun-ai/Step-Audio2 for more information.

Wanlong Liu, Junying Chen, Ke Ji et al.

Retrieval-Augmented Generation (RAG) has emerged as a key paradigm for enhancing large language models (LLMs) by incorporating external knowledge. However, current RAG methods face two limitations: (1) they only cover limited RAG scenarios. (2) They suffer from limited task diversity due to the lack of a general RAG dataset. To address these limitations, we propose RAG-Instruct, a general method for synthesizing diverse and high-quality RAG instruction data based on any source corpus. Our approach leverages (1) five RAG paradigms, which encompass diverse query-document relationships, and (2) instruction simulation, which enhances instruction diversity and quality by utilizing the strengths of existing instruction datasets. Using this method, we construct a 40K instruction dataset from Wikipedia, comprehensively covering diverse RAG scenarios and tasks. Experiments demonstrate that RAG-Instruct effectively enhances LLMs' RAG capabilities, achieving strong zero-shot performance and significantly outperforming various RAG baselines across a diverse set of tasks. RAG-Instruct is publicly available at https://github.com/FreedomIntelligence/RAG-Instruct.

Wanlong Liu, Enqi Zhang, Li Zhou et al.

Recent works have demonstrated the effectiveness of retrieval augmentation in the Event Argument Extraction (EAE) task. However, existing retrieval-based EAE methods have two main limitations: (1) input length constraints and (2) the gap between the retriever and the inference model. These issues limit the diversity and quality of the retrieved information. In this paper, we propose a Compressive Memory-based Retrieval (CMR) mechanism for EAE, which addresses the two limitations mentioned above. Our compressive memory, designed as a dynamic matrix that effectively caches retrieved information and supports continuous updates, overcomes the limitations of the input length. Additionally, after pre-loading all candidate demonstrations into the compressive memory, the model further retrieves and filters relevant information from memory based on the input query, bridging the gap between the retriever and the inference model. Extensive experiments show that our method achieves new state-of-the-art performance on three public datasets (RAMS, WikiEvents, ACE05), significantly outperforming existing retrieval-based EAE methods.

Dayang Wang, Yongshun Xu, Shuo Han et al.

Low-dose computed tomography (LDCT) offers reduced X-ray radiation exposure but at the cost of compromised image quality, characterized by increased noise and artifacts. Recently, transformer models emerged as a promising avenue to enhance LDCT image quality. However, the success of such models relies on a large amount of paired noisy and clean images, which are often scarce in clinical settings. In the fields of computer vision and natural language processing, masked autoencoders (MAE) have been recognized as an effective label-free self-pretraining method for transformers, due to their exceptional feature representation ability. However, the original pretraining and fine-tuning design fails to work in low-level vision tasks like denoising. In response to this challenge, we redesign the classical encoder-decoder learning model and facilitate a simple yet effective low-level vision MAE, referred to as LoMAE, tailored to address the LDCT denoising problem. Moreover, we introduce an MAE-GradCAM method to shed light on the latent learning mechanisms of the MAE/LoMAE. Additionally, we explore the LoMAE's robustness and generability across a variety of noise levels. Experiments results show that the proposed LoMAE can enhance the transformer's denoising performance and greatly relieve the dependence on the ground truth clean data. It also demonstrates remarkable robustness and generalizability over a spectrum of noise levels.

Peiyuan Chen, Zecheng Zhang, Yiping Dong et al.

Visual Question Answering (VQA) is a challenging task that requires systems to provide accurate answers to questions based on image content. Current VQA models struggle with complex questions due to limitations in capturing and integrating multimodal information effectively. To address these challenges, we propose the Rank VQA model, which leverages a ranking-inspired hybrid training strategy to enhance VQA performance. The Rank VQA model integrates high-quality visual features extracted using the Faster R-CNN model and rich semantic text features obtained from a pre-trained BERT model. These features are fused through a sophisticated multimodal fusion technique employing multi-head self-attention mechanisms. Additionally, a ranking learning module is incorporated to optimize the relative ranking of answers, thus improving answer accuracy. The hybrid training strategy combines classification and ranking losses, enhancing the model's generalization ability and robustness across diverse datasets. Experimental results demonstrate the effectiveness of the Rank VQA model. Our model significantly outperforms existing state-of-the-art models on standard VQA datasets, including VQA v2.0 and COCO-QA, in terms of both accuracy and Mean Reciprocal Rank (MRR). The superior performance of Rank VQA is evident in its ability to handle complex questions that require understanding nuanced details and making sophisticated inferences from the image and text. This work highlights the effectiveness of a ranking-based hybrid training strategy in improving VQA performance and lays the groundwork for further research in multimodal learning methods.

Li Zhou, Wenyu Chen, Dingyi Zeng et al.

In the field of natural language understanding, the intersection of neural models and graph meaning representations (GMRs) remains a compelling area of research. Despite the growing interest, a critical gap persists in understanding the exact influence of GMRs, particularly concerning relation extraction tasks. Addressing this, we introduce DAGNN-plus, a simple and parameter-efficient neural architecture designed to decouple contextual representation learning from structural information propagation. Coupled with various sequence encoders and GMRs, this architecture provides a foundation for systematic experimentation on two English and two Chinese datasets. Our empirical analysis utilizes four different graph formalisms and nine parsers. The results yield a nuanced understanding of GMRs, showing improvements in three out of the four datasets, particularly favoring English over Chinese due to highly accurate parsers. Interestingly, GMRs appear less effective in literary-domain datasets compared to general-domain datasets. These findings lay the groundwork for better-informed design of GMRs and parsers to improve relation classification, which is expected to tangibly impact the future trajectory of natural language understanding research.

Wentao Long, Yunfei Zhang, Chenyi Li et al.

Formal theorem-proving benchmarks enable mechanically verifiable evaluation of mathematical reasoning in large language models. However, existing benchmarks mainly focus on Olympiad-style problems and algebraic domains, leaving computational and applied mathematics underrepresented. We introduce CAM-Bench, a Lean 4 theorem-proving benchmark of 1,000 Lean proof targets in computational and applied mathematics, with coverage spanning optimization, numerical linear algebra, and numerical analysis. These problems are adapted from textbook exercises and often depend on locally introduced definitions, notation, algorithms, and elementary results. To construct CAM-Bench, we develop a dependency-recovery pipeline that reconstructs the local textbook context needed to state each problem faithfully. It then normalizes each problem into a standalone informal theorem and translates it into a Lean target. We validate the resulting formal problems through Lean compilation and semantic review, checking both formal correctness and semantic alignment with the original exercises. For each problem, we release the raw exercise, recovered context, normalized informal theorem, and final Lean target. CAM-Bench complements existing formal mathematics benchmarks by targeting applied mathematics problems that rely on textbook concepts and elementary theorems, many of which are not directly available as standard Mathlib4 lemmas. We evaluate widely used large language models and formalization agents on CAM-Bench, and analyze common failure modes in tracking local assumptions, applying elementary results, decomposing proofs, and maintaining long-horizon control in Lean.

Yuning Gong, Yifei Liu, Yifan Zhan et al.

Neural rendering, particularly 3D Gaussian Splatting (3DGS), has evolved rapidly and become a key component for building world models. However, existing viewer solutions remain fragmented, heavy, or constrained by legacy pipelines, resulting in high deployment friction and limited support for dynamic content and generative models. In this work, we present Visionary, an open, web-native platform for real-time various Gaussian Splatting and meshes rendering. Built on an efficient WebGPU renderer with per-frame ONNX inference, Visionary enables dynamic neural processing while maintaining a lightweight, "click-to-run" browser experience. It introduces a standardized Gaussian Generator contract, which not only supports standard 3DGS rendering but also allows plug-and-play algorithms to generate or update Gaussians each frame. Such inference also enables us to apply feedforward generative post-processing. The platform further offers a plug in three.js library with a concise TypeScript API for seamless integration into existing web applications. Experiments show that, under identical 3DGS assets, Visionary achieves superior rendering efficiency compared to current Web viewers due to GPU-based primitive sorting. It already supports multiple variants, including MLP-based 3DGS, 4DGS, neural avatars, and style transformation or enhancement networks. By unifying inference and rendering directly in the browser, Visionary significantly lowers the barrier to reproduction, comparison, and deployment of 3DGS-family methods, serving as a unified World Model Carrier for both reconstructive and generative paradigms.

Wanlong Liu, Dingyi Zeng, Li Zhou et al.

Document-level event argument extraction is a crucial yet challenging task within the field of information extraction. Current mainstream approaches primarily focus on the information interaction between event triggers and their arguments, facing two limitations: insufficient context interaction and the ignorance of event correlations. Here, we introduce a novel framework named CARLG (Contextual Aggregation of clues and Role-based Latent Guidance), comprising two innovative components: the Contextual Clues Aggregation (CCA) and the Role-based Latent Information Guidance (RLIG). The CCA module leverages the attention weights derived from a pre-trained encoder to adaptively assimilates broader contextual information, while the RLIG module aims to capture the semantic correlations among event roles. We then instantiate the CARLG framework into two variants based on two types of current mainstream EAE approaches. Notably, our CARLG framework introduces less than 1% new parameters yet significantly improving the performance. Comprehensive experiments across the RAMS, WikiEvents, and MLEE datasets confirm the superiority of CARLG, showing significant superiority in terms of both performance and inference speed compared to major benchmarks. Further analyses demonstrate the effectiveness of the proposed modules.

Li Zhou, Xu Yuan, Zenghui Sun et al.

Large Multimodal Models (LMMs) have achieved significant progress by extending large language models. Building on this progress, the latest developments in LMMs demonstrate the ability to generate dense pixel-wise segmentation through the integration of segmentation models.Despite the innovations, the textual responses and segmentation masks of existing works remain at the instance level, showing limited ability to perform fine-grained understanding and segmentation even provided with detailed textual cues.To overcome this limitation, we introduce a Multi-Granularity Large Multimodal Model (MGLMM), which is capable of seamlessly adjusting the granularity of Segmentation and Captioning (SegCap) following user instructions, from panoptic SegCap to fine-grained SegCap. We name such a new task Multi-Granularity Segmentation and Captioning (MGSC). Observing the lack of a benchmark for model training and evaluation over the MGSC task, we establish a benchmark with aligned masks and captions in multi-granularity using our customized automated annotation pipeline. This benchmark comprises 10K images and more than 30K image-question pairs. We will release our dataset along with the implementation of our automated dataset annotation pipeline for further research.Besides, we propose a novel unified SegCap data format to unify heterogeneous segmentation datasets; it effectively facilitates learning to associate object concepts with visual features during multi-task training. Extensive experiments demonstrate that our MGLMM excels at tackling more than eight downstream tasks and achieves state-of-the-art performance in MGSC, GCG, image captioning, referring segmentation, multiple and empty segmentation, and reasoning segmentation tasks. The great performance and versatility of MGLMM underscore its potential impact on advancing multimodal research.

Hao Guan, Li Zhou

Vision-Language Models have demonstrated strong potential in medical image analysis and disease diagnosis. However, after deployment, their performance may deteriorate when the input data distribution shifts from that observed during development. Detecting such performance degradation is essential for clinical reliability, yet remains challenging for large pre-trained VLMs operating without labeled data. In this study, we investigate performance degradation detection under data shift in a state-of-the-art pathology VLM. We examine both input-level data shift and output-level prediction behavior to understand their respective roles in monitoring model reliability. To facilitate systematic analysis of input data shift, we develop DomainSAT, a lightweight toolbox with a graphical interface that integrates representative shift detection algorithms and enables intuitive exploration of data shift. Our analysis shows that while input data shift detection is effective at identifying distributional changes and providing early diagnostic signals, it does not always correspond to actual performance degradation. Motivated by this observation, we further study output-based monitoring and introduce a label-free, confidence-based degradation indicator that directly captures changes in model prediction confidence. We find that this indicator exhibits a close relationship with performance degradation and serves as an effective complement to input shift detection. Experiments on a large-scale pathology dataset for tumor classification demonstrate that combining input data shift detection and output confidence-based indicators enables more reliable detection and interpretation of performance degradation in VLMs under data shift. These findings provide a practical and complementary framework for monitoring the reliability of foundation models in digital pathology.

Wenyan Li, Xinyu Zhang, Jiaang Li et al.

Food is a rich and varied dimension of cultural heritage, crucial to both individuals and social groups. To bridge the gap in the literature on the often-overlooked regional diversity in this domain, we introduce FoodieQA, a manually curated, fine-grained image-text dataset capturing the intricate features of food cultures across various regions in China. We evaluate vision-language Models (VLMs) and large language models (LLMs) on newly collected, unseen food images and corresponding questions. FoodieQA comprises three multiple-choice question-answering tasks where models need to answer questions based on multiple images, a single image, and text-only descriptions, respectively. While LLMs excel at text-based question answering, surpassing human accuracy, the open-sourced VLMs still fall short by 41% on multi-image and 21% on single-image VQA tasks, although closed-weights models perform closer to human levels (within 10%). Our findings highlight that understanding food and its cultural implications remains a challenging and under-explored direction.

Tenggan Zhang, Xinjie Zhang, Jinming Zhao et al.

Understanding the reason for emotional support response is crucial for establishing connections between users and emotional support dialogue systems. Previous works mostly focus on generating better responses but ignore interpretability, which is extremely important for constructing reliable dialogue systems. To empower the system with better interpretability, we propose an emotional support response generation scheme, named $\textbf{E}$motion-Focused and $\textbf{S}$trategy-Driven $\textbf{C}$hain-$\textbf{o}$f-$\textbf{T}$hought ($\textbf{ESCoT}$), mimicking the process of $\textit{identifying}$, $\textit{understanding}$, and $\textit{regulating}$ emotions. Specially, we construct a new dataset with ESCoT in two steps: (1) $\textit{Dialogue Generation}$ where we first generate diverse conversation situations, then enhance dialogue generation using richer emotional support strategies based on these situations; (2) $\textit{Chain Supplement}$ where we focus on supplementing selected dialogues with elements such as emotion, stimuli, appraisal, and strategy reason, forming the manually verified chains. Additionally, we further develop a model to generate dialogue responses with better interpretability. We also conduct extensive experiments and human evaluations to validate the effectiveness of the proposed ESCoT and generated dialogue responses. Our data and code are available at $\href{https://github.com/TeigenZhang/ESCoT}{https://github.com/TeigenZhang/ESCoT}$.

Li Zhou, Kevin Small

Multi-domain dialogue state tracking (DST) is a critical component for conversational AI systems. The domain ontology (i.e., specification of domains, slots, and values) of a conversational AI system is generally incomplete, making the capability for DST models to generalize to new slots, values, and domains during inference imperative. In this paper, we propose to model multi-domain DST as a question answering problem, referred to as Dialogue State Tracking via Question Answering (DSTQA). Within DSTQA, each turn generates a question asking for the value of a (domain, slot) pair, thus making it naturally extensible to unseen domains, slots, and values. Additionally, we use a dynamically-evolving knowledge graph to explicitly learn relationships between (domain, slot) pairs. Our model has a 5.80% and 12.21% relative improvement over the current state-of-the-art model on MultiWOZ 2.0 and MultiWOZ 2.1 datasets, respectively. Additionally, our model consistently outperforms the state-of-the-art model in domain adaptation settings. (Code is released at https://github.com/alexa/dstqa )

Ruisong Zhou, Haijun Zou, Li Zhou et al.

Efficient scheduling of directed acyclic graphs (DAGs) in heterogeneous environments is challenging due to resource capacities and dependencies. In practice, the need for adaptability across environments with varying resource pools and task types, alongside rapid schedule generation, complicates these challenges. We propose WeCAN, an end-to-end reinforcement learning framework for heterogeneous DAG scheduling that addresses task--pool compatibility coefficients and generation-induced optimality gaps. It adopts a two-stage single-pass design: a single forward pass produces task--pool scores and global parameters, followed by a generation map that constructs schedules without repeated network calls. Its weighted cross-attention encoder models task--pool interactions gated by compatibility coefficients, and is size-agnostic to environment fluctuations. Moreover, widely used list-scheduling maps can incur generation-induced optimality gaps from restricted reachability. We introduce an order-space analysis that characterizes the reachable set of generation maps via feasible schedule orders, explains the mechanism behind generation-induced gaps, and yields sufficient conditions for gap elimination. Guided by these conditions, we design a skip-extended realization with an analytically parameterized decreasing skip rule, which enlarges the reachable order set while preserving single-pass efficiency. Experiments on computation graphs and real-world TPC-H DAGs demonstrate improved makespan over strong baselines, with inference time comparable to classical heuristics and faster than multi-round neural schedulers.

Li Zhou, Hao Jiang, Junjie Li et al.

Achieving precise and controllable emotional expression is crucial for producing natural and context-appropriate speech in text-to-speech (TTS) synthesis. However, many emotion-aware TTS systems, including large language model (LLM)-based designs, rely on scaling fixed emotion embeddings or external guidance, limiting their ability to model emotion-specific latent characteristics. To address this gap, we present EmoShift, a lightweight activation-steering framework incorporating a EmoSteer layer, which learns a steering vector for each target emotion in the output embedding space to capture its latent offset and maintain stable, appropriate expression across utterances and categories. With only 10M trainable parameters,less than 1/30 of full fine-tuning, EmoShift outperforms zero-shot and fully fine-tuned baselines in objective and subjective evaluations, enhancing emotional expressiveness while preserving naturalness and speaker similarity. Further analysis confirms the proposed EmoSteer layer's effectiveness and reveals its potential for controllable emotional intensity in speech synthesis.

Wanlong Liu, Li Zhou, Dingyi Zeng et al.

Recent mainstream event argument extraction methods process each event in isolation, resulting in inefficient inference and ignoring the correlations among multiple events. To address these limitations, here we propose a multiple-event argument extraction model DEEIA (Dependency-guided Encoding and Event-specific Information Aggregation), capable of extracting arguments from all events within a document simultaneouslyThe proposed DEEIA model employs a multi-event prompt mechanism, comprising DE and EIA modules. The DE module is designed to improve the correlation between prompts and their corresponding event contexts, whereas the EIA module provides event-specific information to improve contextual understanding. Extensive experiments show that our method achieves new state-of-the-art performance on four public datasets (RAMS, WikiEvents, MLEE, and ACE05), while significantly saving the inference time compared to the baselines. Further analyses demonstrate the effectiveness of the proposed modules.

GaoXiang Zhao, Li Zhou, XiaoQiang Wang

Long time series forecasting aims to utilize historical information to forecast future states over extended horizons. Traditional RNN-based series forecasting methods struggle to effectively address long-term dependencies and gradient issues in long time series problems. Recently, SegRNN has emerged as a leading RNN-based model tailored for long-term series forecasting, demonstrating state-of-the-art performance while maintaining a streamlined architecture through innovative segmentation and parallel decoding techniques. Nevertheless, SegRNN has several limitations: its fixed segmentation disrupts data continuity and fails to effectively leverage information across different segments, the segmentation strategy employed by SegRNN does not fundamentally address the issue of information loss within the recurrent structure. To address these issues, we propose the ISMRNN method with three key enhancements: we introduce an implicit segmentation structure to decompose the time series and map it to segmented hidden states, resulting in denser information exchange during the segmentation phase. Additionally, we incorporate residual structures in the encoding layer to mitigate information loss within the recurrent structure. To extract information more effectively, we further integrate the Mamba architecture to enhance time series information extraction. Experiments on several real-world long time series forecasting datasets demonstrate that our model surpasses the performance of current state-of-the-art models.

Shizhe Yuan, Li Zhou

With the advancement of artificial intelligence, 3D human pose estimation-based systems for sports training and posture correction have gained significant attention in adolescent sports. However, existing methods face challenges in handling complex movements, providing real-time feedback, and accommodating diverse postures, particularly with occlusions, rapid movements, and the resource constraints of Internet of Things (IoT) devices, making it difficult to balance accuracy and real-time performance. To address these issues, we propose GTA-Net, an intelligent system for posture correction and real-time feedback in adolescent sports, integrated within an IoT-enabled environment. This model enhances pose estimation in dynamic scenes by incorporating Graph Convolutional Networks (GCN), Temporal Convolutional Networks (TCN), and Hierarchical Attention mechanisms, achieving real-time correction through IoT devices. Experimental results show GTA-Net's superior performance on Human3.6M, HumanEva-I, and MPI-INF-3DHP datasets, with Mean Per Joint Position Error (MPJPE) values of 32.2mm, 15.0mm, and 48.0mm, respectively, significantly outperforming existing methods. The model also demonstrates strong robustness in complex scenarios, maintaining high accuracy even with occlusions and rapid movements. This system enhances real-time posture correction and offers broad applications in intelligent sports and health management.

Zihao Cheng, Li Zhou, Feng Jiang et al.

The rapid development of large language models (LLMs), like ChatGPT, has resulted in the widespread presence of LLM-generated content on social media platforms, raising concerns about misinformation, data biases, and privacy violations, which can undermine trust in online discourse. While detecting LLM-generated content is crucial for mitigating these risks, current methods often focus on binary classification, failing to address the complexities of real-world scenarios like human-LLM collaboration. To move beyond binary classification and address these challenges, we propose a new paradigm for detecting LLM-generated content. This approach introduces two novel tasks: LLM Role Recognition (LLM-RR), a multi-class classification task that identifies specific roles of LLM in content generation, and LLM Influence Measurement (LLM-IM), a regression task that quantifies the extent of LLM involvement in content creation. To support these tasks, we propose LLMDetect, a benchmark designed to evaluate detectors' performance on these new tasks. LLMDetect includes the Hybrid News Detection Corpus (HNDC) for training detectors, as well as DetectEval, a comprehensive evaluation suite that considers five distinct cross-context variations and two multi-intensity variations within the same LLM role. This allows for a thorough assessment of detectors' generalization and robustness across diverse contexts. Our empirical validation of 10 baseline detection methods demonstrates that fine-tuned PLM-based models consistently outperform others on both tasks, while advanced LLMs face challenges in accurately detecting their own generated content. Our experimental results and analysis offer insights for developing more effective detection models for LLM-generated content. This research enhances the understanding of LLM-generated content and establishes a foundation for more nuanced detection methodologies.

Huadong Pang, Li Zhou, Yiping Dong et al.

In the healthcare sector, the application of deep learning technologies has revolutionized data analysis and disease forecasting. This is particularly evident in the field of diabetes, where the deep analysis of Electronic Health Records (EHR) has unlocked new opportunities for early detection and effective intervention strategies. Our research presents an innovative model that synergizes the capabilities of Bidirectional Long Short-Term Memory Networks-Conditional Random Field (BiLSTM-CRF) with a fusion of XGBoost and Logistic Regression. This model is designed to enhance the accuracy of diabetes risk prediction by conducting an in-depth analysis of electronic medical records data. The first phase of our approach involves employing BiLSTM-CRF to delve into the temporal characteristics and latent patterns present in EHR data. This method effectively uncovers the progression trends of diabetes, which are often hidden in the complex data structures of medical records. The second phase leverages the combined strength of XGBoost and Logistic Regression to classify these extracted features and evaluate associated risks. This dual approach facilitates a more nuanced and precise prediction of diabetes, outperforming traditional models, particularly in handling multifaceted and nonlinear medical datasets. Our research demonstrates a notable advancement in diabetes prediction over traditional methods, showcasing the effectiveness of our combined BiLSTM-CRF, XGBoost, and Logistic Regression model. This study highlights the value of data-driven strategies in clinical decision-making, equipping healthcare professionals with precise tools for early detection and intervention. By enabling personalized treatment and timely care, our approach signifies progress in incorporating advanced analytics in healthcare, potentially improving outcomes for diabetes and other chronic conditions.

Fengyi Shen, Li Zhou, Kagan Kucukaytekin et al.

Data generation is recognized as a potent strategy for unsupervised domain adaptation (UDA) pertaining semantic segmentation in adverse weathers. Nevertheless, these adverse weather scenarios encompass multiple possibilities, and high-fidelity data synthesis with controllable weather is under-researched in previous UDA works. The recent strides in large-scale text-to-image diffusion models (DM) have ushered in a novel avenue for research, enabling the generation of realistic images conditioned on semantic labels. This capability proves instrumental for cross-domain data synthesis from source to target domain owing to their shared label space. Thus, source domain labels can be paired with those generated pseudo target data for training UDA. However, from the UDA perspective, there exists several challenges for DM training: (i) ground-truth labels from target domain are missing; (ii) the prompt generator may produce vague or noisy descriptions of images from adverse weathers; (iii) existing arts often struggle to well handle the complex scene structure and geometry of urban scenes when conditioned only on semantic labels. To tackle the above issues, we propose ControlUDA, a diffusion-assisted framework tailored for UDA segmentation under adverse weather conditions. It first leverages target prior from a pre-trained segmentor for tuning the DM, compensating the missing target domain labels; It also contains UDAControlNet, a condition-fused multi-scale and prompt-enhanced network targeted at high-fidelity data generation in adverse weathers. Training UDA with our generated data brings the model performances to a new milestone (72.0 mIoU) on the popular Cityscapes-to-ACDC benchmark for adverse weathers. Furthermore, ControlUDA helps to achieve good model generalizability on unseen data.

Hao Guan, David Bates, Li Zhou

Artificial intelligence (AI) is increasingly integrated into modern healthcare, offering powerful support for clinical decision-making. However, in real-world settings, AI systems may experience performance degradation over time, due to factors such as shifting data distributions, changes in patient characteristics, evolving clinical protocols, and variations in data quality. These factors can compromise model reliability, posing safety concerns and increasing the likelihood of inaccurate predictions or adverse outcomes. This review presents a forward-looking perspective on monitoring and maintaining the "health" of AI systems in healthcare. We highlight the urgent need for continuous performance monitoring, early degradation detection, and effective self-correction mechanisms. The paper begins by reviewing common causes of performance degradation at both data and model levels. We then summarize key techniques for detecting data and model drift, followed by an in-depth look at root cause analysis. Correction strategies are further reviewed, ranging from model retraining to test-time adaptation. Our survey spans both traditional machine learning models and state-of-the-art large language models (LLMs), offering insights into their strengths and limitations. Finally, we discuss ongoing technical challenges and propose future research directions. This work aims to guide the development of reliable, robust medical AI systems capable of sustaining safe, long-term deployment in dynamic clinical settings.

Kuang Wang, Xianfei Li, Shenghao Yang et al.

User simulators are crucial for replicating human interactions with dialogue systems, supporting both collaborative training and automatic evaluation, especially for large language models (LLMs). However, current role-playing methods face challenges such as a lack of utterance-level authenticity and user-level diversity, often hindered by role confusion and dependence on predefined profiles of well-known figures. In contrast, direct simulation focuses solely on text, neglecting implicit user traits like personality and conversation-level consistency. To address these issues, we introduce the User Simulator with Implicit Profiles (USP), a framework that infers implicit user profiles from human-machine interactions to simulate personalized and realistic dialogues. We first develop an LLM-driven extractor with a comprehensive profile schema, then refine the simulation using conditional supervised fine-tuning and reinforcement learning with cycle consistency, optimizing at both the utterance and conversation levels. Finally, a diverse profile sampler captures the distribution of real-world user profiles. Experimental results show that USP outperforms strong baselines in terms of authenticity and diversity while maintaining comparable consistency. Additionally, using USP to evaluate LLM on dynamic multi-turn aligns well with mainstream benchmarks, demonstrating its effectiveness in real-world applications.

Tianyi Lyu, Dian Gu, Peiyuan Chen et al.

This study introduces a method for efficiently detecting objects within 3D point clouds using convolutional neural networks (CNNs). Our approach adopts a unique feature-centric voting mechanism to construct convolutional layers that capitalize on the typical sparsity observed in input data. We explore the trade-off between accuracy and speed across diverse network architectures and advocate for integrating an $\mathcal{L}_1$ penalty on filter activations to augment sparsity within intermediate layers. This research pioneers the proposal of sparse convolutional layers combined with $\mathcal{L}_1$ regularization to effectively handle large-scale 3D data processing. Our method's efficacy is demonstrated on the MVTec 3D-AD object detection benchmark. The Vote3Deep models, with just three layers, outperform the previous state-of-the-art in both laser-only approaches and combined laser-vision methods. Additionally, they maintain competitive processing speeds. This underscores our approach's capability to substantially enhance detection performance while ensuring computational efficiency suitable for real-time applications.

Xinqi Liu, Li Zhou, Zikun Zhou et al.

The vision-language tracking task aims to perform object tracking based on various modality references. Existing Transformer-based vision-language tracking methods have made remarkable progress by leveraging the global modeling ability of self-attention. However, current approaches still face challenges in effectively exploiting the temporal information and dynamically updating reference features during tracking. Recently, the State Space Model (SSM), known as Mamba, has shown astonishing ability in efficient long-sequence modeling. Particularly, its state space evolving process demonstrates promising capabilities in memorizing multimodal temporal information with linear complexity. Witnessing its success, we propose a Mamba-based vision-language tracking model to exploit its state space evolving ability in temporal space for robust multimodal tracking, dubbed MambaVLT. In particular, our approach mainly integrates a time-evolving hybrid state space block and a selective locality enhancement block, to capture contextual information for multimodal modeling and adaptive reference feature update. Besides, we introduce a modality-selection module that dynamically adjusts the weighting between visual and language references, mitigating potential ambiguities from either reference type. Extensive experimental results show that our method performs favorably against state-of-the-art trackers across diverse benchmarks.

Xiaowei Tang, Bin Long, Li Zhou

This research focuses on real-time monitoring and analysis of track and field athletes, addressing the limitations of traditional monitoring systems in terms of real-time performance and accuracy. We propose an IoT-optimized system that integrates edge computing and deep learning algorithms. Traditional systems often experience delays and reduced accuracy when handling complex motion data, whereas our method, by incorporating a SAC-optimized deep learning model within the IoT architecture, achieves efficient motion recognition and real-time feedback. Experimental results show that this system significantly outperforms traditional methods in response time, data processing accuracy, and energy efficiency, particularly excelling in complex track and field events. This research not only enhances the precision and efficiency of athlete monitoring but also provides new technical support and application prospects for sports science research.

Li Zhou, Ruijie Zhang, Xunlian Dai et al.

While the widespread use of Large Language Models (LLMs) brings convenience, it also raises concerns about the credibility of academic research and scholarly processes. To better understand these dynamics, we evaluate the penetration of LLMs across academic workflows from multiple perspectives and dimensions, providing compelling evidence of their growing influence. We propose a framework with two components: \texttt{ScholarLens}, a curated dataset of human- and LLM-generated content across scholarly writing and peer review for multi-perspective evaluation, and \texttt{LLMetrica}, a tool for assessing LLM penetration using rule-based metrics and model-based detectors for multi-dimensional evaluation. Our experiments demonstrate the effectiveness of \texttt{LLMetrica}, revealing the increasing role of LLMs in scholarly processes. These findings emphasize the need for transparency, accountability, and ethical practices in LLM usage to maintain academic credibility.

Janick Weberpals, Pamela A. Shaw, Kueiyu Joshua Lin et al.

Multiple imputation (MI) models can be improved by including auxiliary covariates (AC), but their performance in high-dimensional data is not well understood. We aimed to develop and compare high-dimensional MI (HDMI) approaches using structured and natural language processing (NLP)-derived AC in studies with partially observed confounders. We conducted a plasmode simulation study using data from opioid vs. non-steroidal anti-inflammatory drug (NSAID) initiators (X) with observed serum creatinine labs (Z2) and time-to-acute kidney injury as outcome. We simulated 100 cohorts with a null treatment effect, including X, Z2, atrial fibrillation (U), and 13 other investigator-derived confounders (Z1) in the outcome generation. We then imposed missingness (MZ2) on 50% of Z2 measurements as a function of Z2 and U and created different HDMI candidate AC using structured and NLP-derived features. We mimicked scenarios where U was unobserved by omitting it from all AC candidate sets. Using LASSO, we data-adaptively selected HDMI covariates associated with Z2 and MZ2 for MI, and with U to include in propensity score models. The treatment effect was estimated following propensity score matching in MI datasets and we benchmarked HDMI approaches against a baseline imputation and complete case analysis with Z1 only. HDMI using claims data showed the lowest bias (0.072). Combining claims and sentence embeddings led to an improvement in the efficiency displaying the lowest root-mean-squared-error (0.173) and coverage (94%). NLP-derived AC alone did not perform better than baseline MI. HDMI approaches may decrease bias in studies with partially observed confounders where missingness depends on unobserved factors.

Li Zhou, Wenyu Chen, Yong Cao et al.

While Transformer-based pre-trained language models and their variants exhibit strong semantic representation capabilities, the question of comprehending the information gain derived from the additional components of PLMs remains an open question in this field. Motivated by recent efforts that prove Multilayer-Perceptrons (MLPs) modules achieving robust structural capture capabilities, even outperforming Graph Neural Networks (GNNs), this paper aims to quantify whether simple MLPs can further enhance the already potent ability of PLMs to capture linguistic information. Specifically, we design a simple yet effective probing framework containing MLPs components based on BERT structure and conduct extensive experiments encompassing 10 probing tasks spanning three distinct linguistic levels. The experimental results demonstrate that MLPs can indeed enhance the comprehension of linguistic structure by PLMs. Our research provides interpretable and valuable insights into crafting variations of PLMs utilizing MLPs for tasks that emphasize diverse linguistic structures.

Bo Liu, Runlong Li, Li Zhou et al.

This paper proposes a Diffusion Model-Optimized Neural Radiance Field (DT-NeRF) method, aimed at enhancing detail recovery and multi-view consistency in 3D scene reconstruction. By combining diffusion models with Transformers, DT-NeRF effectively restores details under sparse viewpoints and maintains high accuracy in complex geometric scenes. Experimental results demonstrate that DT-NeRF significantly outperforms traditional NeRF and other state-of-the-art methods on the Matterport3D and ShapeNet datasets, particularly in metrics such as PSNR, SSIM, Chamfer Distance, and Fidelity. Ablation experiments further confirm the critical role of the diffusion and Transformer modules in the model's performance, with the removal of either module leading to a decline in performance. The design of DT-NeRF showcases the synergistic effect between modules, providing an efficient and accurate solution for 3D scene reconstruction. Future research may focus on further optimizing the model, exploring more advanced generative models and network architectures to enhance its performance in large-scale dynamic scenes.

Yuhao Du, Qianwei Huang, Guo Zhu et al.

The rapid advancement of speech-to-speech (S2S) large language models (LLMs) has significantly improved real-time spoken interaction. However, current evaluation frameworks remain inadequate for assessing performance in complex, multi-turn dialogues. To address this, we introduce MTalk-Bench, a multi-turn S2S benchmark covering three core dimensions: Semantic Information, Paralinguistic Information, and Ambient Sound. Each dimension includes nine realistic scenarios, along with targeted tasks to assess specific capabilities such as reasoning. Our dual-method evaluation framework combines Arena-style evaluation (pairwise comparison) and Rubrics-based evaluation (absolute scoring) for relative and absolute assessment. The benchmark includes both model and human outputs, evaluated by human evaluators and LLMs. Experimental results reveal two sets of findings. Overall performance of S2S LLMs: (1) models excel at semantic information processing yet underperform on paralinguistic information and ambient sounds perception; (2) models typically regain coherence by increasing response length, sacrificing efficiency in multi-turn dialogues; (3) modality-aware, task-specific designs outperform brute scaling. Evaluation framework and reliability: (1) Arena and Rubrics yield consistent, complementary rankings, but reliable distinctions emerge only when performance gaps are large; (2) LLM-as-a-judge aligns with humans when gaps are clear or criteria explicit, but exhibits position and length biases and is reliable on nonverbal evaluation only with text annotations. These results highlight current limitations in S2S evaluation and the need for more robust, speech-aware assessment frameworks.

Kaidong Feng, Zhuoxuan Huang, Huizhong Guo et al.

Fashion understanding requires both visual perception and expert-level reasoning about style, occasion, compatibility, and outfit rationale. However, existing fashion datasets remain fragmented and task-specific, often focusing on item attributes, outfit co-occurrence, or weak textual supervision, and thus provide limited support for holistic outfit understanding. In this paper, we introduce FashionStylist, an expert-annotated benchmark for holistic and expert-level fashion understanding. Constructed through a dedicated fashion-expert annotation pipeline, FashionStylist provides professionally grounded annotations at both the item and outfit levels. It supports three representative tasks: outfit-to-item grounding, outfit completion, and outfit evaluation. These tasks cover realistic item recovery from complex outfits with layering and accessories, compatibility-aware composition beyond co-occurrence matching, and expert-level assessment of style, season, occasion, and overall coherence. Experimental results show that FashionStylist serves not only as a unified benchmark for multiple fashion tasks, but also as an effective training resource for improving grounding, completion, and outfit-level semantic evaluation in MLLM-based fashion systems.

Xin Dong, Shichao Dong, Jin Wang et al.

Hallucinations in large vision-language models (LVLMs) pose significant challenges for real-world applications, as LVLMs may generate responses that appear plausible yet remain inconsistent with the associated visual content. This issue rarely occurs in human cognition. We argue that this discrepancy arises from humans' ability to effectively leverage multimodal interaction information in data samples. Specifically, humans typically first gather multimodal information, analyze the interactions across modalities for understanding, and then express their understanding through language. Motivated by this observation, we conduct extensive experiments on popular LVLMs and obtained insights that surprisingly reveal human-like, though less pronounced, cognitive behavior of LVLMs on multimodal samples. Building on these findings, we further propose \textbf{INTER}: \textbf{Inter}action Guidance Sampling, a novel training-free algorithm that mitigate hallucinations without requiring additional data. Specifically, INTER explicitly guides LVLMs to effectively reapply their understanding of multimodal interaction information when generating responses, thereby reducing potential hallucinations. On six benchmarks including VQA and image captioning tasks, INTER achieves an average improvement of up to 3.4\% on five LVLMs compared to the state-of-the-art decoding strategy. The code will be released when the paper is accepted.

Xunlian Dai, Li Zhou, Benyou Wang et al.

The human-centered word association test (WAT) serves as a cognitive proxy, revealing sociocultural variations through culturally shared semantic expectations and implicit linguistic patterns shaped by lived experiences. We extend this test into an LLM-adaptive, free-relation task to assess the alignment of large language models (LLMs) with cross-cultural cognition. To address culture preference, we propose CultureSteer, an innovative approach that moves beyond superficial cultural prompting by embedding cultural-specific semantic associations directly within the model's internal representation space. Experiments show that current LLMs exhibit significant bias toward Western (notably American) schemas at the word association level. In contrast, our model substantially improves cross-cultural alignment, capturing diverse semantic associations. Further validation on culture-sensitive downstream tasks confirms its efficacy in fostering cognitive alignment across cultures. This work contributes a novel methodological paradigm for enhancing cultural awareness in LLMs, advancing the development of more inclusive language technologies.