Qi Jia

Yuqing Liu, Qi Jia, Jian Zhang et al.

Bit-depth expansion (BDE) is one of the emerging technologies to display high bit-depth (HBD) image from low bit-depth (LBD) source. Existing BDE methods have no unified solution for various BDE situations, and directly learn a mapping for each pixel from LBD image to the desired value in HBD image, which may change the given high-order bits and lead to a huge deviation from the ground truth. In this paper, we design a bit restoration network (BRNet) to learn a weight for each pixel, which indicates the ratio of the replenished value within a rational range, invoking an accurate solution without modifying the given high-order bit information. To make the network adaptive for any bit-depth degradation, we investigate the issue in an optimization perspective and train the network under progressive training strategy for better performance. Moreover, we employ Wasserstein distance as a visual quality indicator to evaluate the difference of color distribution between restored image and the ground truth. Experimental results show our method can restore colorful images with fewer artifacts and false contours, and outperforms state-of-the-art methods with higher PSNR/SSIM results and lower Wasserstein distance. The source code will be made available at https://github.com/yuqing-liu-dut/bit-depth-expansion

Yuqing Liu, Qi Jia, Shanshe Wang et al.

Image super-resolution (SR) has been widely investigated in recent years. However, it is challenging to fairly estimate the performance of various SR methods, as the lack of reliable and accurate criteria for the perceptual quality. Existing metrics concentrate on the specific kind of degradation without distinguishing the visual sensitive areas, which have no ability to describe the diverse SR degeneration situations in both low-level textural and high-level perceptual information. In this paper, we focus on the textural and perceptual degradation of SR images, and design a dual stream network to jointly explore the textural and perceptual information for quality assessment, dubbed TPNet. By mimicking the human vision system (HVS) that pays more attention to the significant image areas, we develop the spatial attention to make the visual sensitive information more distinguishable and utilize feature normalization (F-Norm) to boost the network representation. Experimental results show the TPNet predicts the visual quality score more accurate than other methods and demonstrates better consistency with the human's perspective. The source code will be available at \url{http://github.com/yuqing-liu-dut/NRIQA_SR}

Yunhao Li, Sijing Wu, Zhilin Gao et al.

Large multimodal models (LMMs) have demonstrated outstanding capabilities in various visual perception tasks, which has in turn made the evaluation of LMMs significant. However, the capability of video aesthetic quality assessment, which is a fundamental ability for human, remains underexplored for LMMs. To address this, we introduce VideoAesBench, a comprehensive benchmark for evaluating LMMs' understanding of video aesthetic quality. VideoAesBench has several significant characteristics: (1) Diverse content including 1,804 videos from multiple video sources including user-generated (UGC), AI-generated (AIGC), compressed, robotic-generated (RGC), and game videos. (2) Multiple question formats containing traditional single-choice questions, multi-choice questions, True or False questions, and a novel open-ended questions for video aesthetics description. (3) Holistic video aesthetics dimensions including visual form related questions from 5 aspects, visual style related questions from 4 aspects, and visual affectiveness questions from 3 aspects. Based on VideoAesBench, we benchmark 23 open-source and commercial large multimodal models. Our findings show that current LMMs only contain basic video aesthetics perception ability, their performance remains incomplete and imprecise. We hope our VideoAesBench can be served as a strong testbed and offer insights for explainable video aesthetics assessment. The data will be released on https://github.com/michaelliyunhao/VideoAesBench

Sijing Wu, Yunhao Li, Zicheng Zhang et al.

Recent advances in multimodal large language models (MLLMs) have demonstrated impressive performance on existing low-level vision benchmarks, which primarily focus on generic images. However, their capabilities to perceive and assess portrait images, a domain characterized by distinct structural and perceptual properties, remain largely underexplored. To this end, we introduce Q-Bench-Portrait, the first holistic benchmark specifically designed for portrait image quality perception, comprising 2,765 image-question-answer triplets and featuring (1) diverse portrait image sources, including natural, synthetic distortion, AI-generated, artistic, and computer graphics images; (2) comprehensive quality dimensions, covering technical distortions, AIGC-specific distortions, and aesthetics; and (3) a range of question formats, including single-choice, multiple-choice, true/false, and open-ended questions, at both global and local levels. Based on Q-Bench-Portrait, we evaluate 20 open-source and 5 closed-source MLLMs, revealing that although current models demonstrate some competence in portrait image perception, their performance remains limited and imprecise, with a clear gap relative to human judgments. We hope that the proposed benchmark will foster further research into enhancing the portrait image perception capabilities of both general-purpose and domain-specific MLLMs.

Ye Shen, Dun Pei, Yiqiu Guo et al.

Despite recent advances in understanding and leveraging long-range conversational memory, existing benchmarks still lack systematic evaluation of large language models(LLMs) across diverse memory dimensions, particularly in multi-session settings. In this work, we propose EvolMem, a new benchmark for assessing multi-session memory capabilities of LLMs and agent systems. EvolMem is grounded in cognitive psychology and encompasses both declarative and non-declarative memory, further decomposed into multiple fine-grained abilities. To construct the benchmark, we introduce a hybrid data synthesis framework that consists of topic-initiated generation and narrative-inspired transformations. This framework enables scalable generation of multi-session conversations with controllable complexity, accompanied by sample-specific evaluation guidelines. Extensive evaluation reveals that no LLM consistently outperforms others across all memory dimensions. Moreover, agent memory mechanisms do not necessarily enhance LLMs' capabilities and often exhibit notable efficiency limitations. Data and code will be released at https://github.com/shenye7436/EvolMem.

Siyu Ren, Qi Jia, Kenny Q. Zhu

The quadratic complexity of the attention module makes it gradually become the bulk of compute in Transformer-based LLMs during generation. Moreover, the excessive key-value cache that arises when dealing with long inputs also brings severe issues on memory footprint and inference latency. In this work, we propose a plug-and-play approach that is able to incrementally compress the intermediate activation of a specified span of tokens into compact ones, thereby reducing both memory and computational cost when processing subsequent context. Experiments on both in-domain language modeling and zero-shot open-ended document generation demonstrate the advantage of our approach over sparse attention baselines in terms of fluency, n-gram matching, and semantic similarity. At last, we comprehensively profile the benefit of context compression on improving the system throughout. Code is available at https://github.com/DRSY/KV_Compression.

Yuqing Liu, Qi Jia, Jian Zhang et al.

As a highly ill-posed issue, single image super-resolution (SISR) has been widely investigated in recent years. The main task of SISR is to recover the information loss caused by the degradation procedure. According to the Nyquist sampling theory, the degradation leads to aliasing effect and makes it hard to restore the correct textures from low-resolution (LR) images. In practice, there are correlations and self-similarities among the adjacent patches in the natural images. This paper considers the self-similarity and proposes a hierarchical image super-resolution network (HSRNet) to suppress the influence of aliasing. We consider the SISR issue in the optimization perspective, and propose an iterative solution pattern based on the half-quadratic splitting (HQS) method. To explore the texture with local image prior, we design a hierarchical exploration block (HEB) and progressive increase the receptive field. Furthermore, multi-level spatial attention (MSA) is devised to obtain the relations of adjacent feature and enhance the high-frequency information, which acts as a crucial role for visual experience. Experimental result shows HSRNet achieves better quantitative and visual performance than other works, and remits the aliasing more effectively.

Yixin Guo, Yu Liu, Jianghao Li et al.

Zero-shot human-object interaction (HOI) detector is capable of generalizing to HOI categories even not encountered during training. Inspired by the impressive zero-shot capabilities offered by CLIP, latest methods strive to leverage CLIP embeddings for improving zero-shot HOI detection. However, these embedding-based methods train the classifier on seen classes only, inevitably resulting in seen-unseen confusion for the model during inference. Besides, we find that using prompt-tuning and adapters further increases the gap between seen and unseen accuracy. To tackle this challenge, we present the first generation-based model using CLIP for zero-shot HOI detection, coined HOIGen. It allows to unlock the potential of CLIP for feature generation instead of feature extraction only. To achieve it, we develop a CLIP-injected feature generator in accordance with the generation of human, object and union features. Then, we extract realistic features of seen samples and mix them with synthetic features together, allowing the model to train seen and unseen classes jointly. To enrich the HOI scores, we construct a generative prototype bank in a pairwise HOI recognition branch, and a multi-knowledge prototype bank in an image-wise HOI recognition branch, respectively. Extensive experiments on HICO-DET benchmark demonstrate our HOIGen achieves superior performance for both seen and unseen classes under various zero-shot settings, compared with other top-performing methods. Code is available at: https://github.com/soberguo/HOIGen

Bohan Liang, Zijian Chen, Qi Jia et al.

Stock prediction, a subject closely related to people's investment activities in fully dynamic and live environments, has been widely studied. Current large language models (LLMs) have shown remarkable potential in various domains, exhibiting expert-level performance through advanced reasoning and contextual understanding. In this paper, we introduce PriceSeer, a live, dynamic, and data-uncontaminated benchmark specifically designed for LLMs performing stock prediction tasks. Specifically, PriceSeer includes 110 U.S. stocks from 11 industrial sectors, with each containing 249 historical data points. Our benchmark implements both internal and external information expansion, where LLMs receive extra financial indicators, news, and fake news to perform stock price prediction. We evaluate six cutting-edge LLMs under different prediction horizons, demonstrating their potential in generating investment strategies after obtaining accurate price predictions for different sectors. Additionally, we provide analyses of LLMs' suboptimal performance in long-term predictions, including the vulnerability to fake news and specific industries. The code and evaluation data will be open-sourced at https://github.com/BobLiang2113/PriceSeer.

Qi Jia, Yizhu Liu, Siyu Ren et al.

Abstractive dialogue summarization is to generate a concise and fluent summary covering the salient information in a dialogue among two or more interlocutors. It has attracted great attention in recent years based on the massive emergence of social communication platforms and an urgent requirement for efficient dialogue information understanding and digestion. Different from news or articles in traditional document summarization, dialogues bring unique characteristics and additional challenges, including different language styles and formats, scattered information, flexible discourse structures and unclear topic boundaries. This survey provides a comprehensive investigation on existing work for abstractive dialogue summarization from scenarios, approaches to evaluations. It categorizes the task into two broad categories according to the type of input dialogues, i.e., open-domain and task-oriented, and presents a taxonomy of existing techniques in three directions, namely, injecting dialogue features, designing auxiliary training tasks and using additional data.A list of datasets under different scenarios and widely-accepted evaluation metrics are summarized for completeness. After that, the trends of scenarios and techniques are summarized, together with deep insights on correlations between extensively exploited features and different scenarios. Based on these analyses, we recommend future directions including more controlled and complicated scenarios, technical innovations and comparisons, publicly available datasets in special domains, etc.

Yuting Zheng, Zijian Chen, Qi Jia

Unraveling the hierarchical structure-property relationships is the central challenge of materials science, necessitating the interpretation of data across vast physical scales from micro to macro. Despite the rapid integration of Large Multimodal Models (LMMs) into scientific workflows, existing scientific benchmarks primarily focus on general chart interpretation or isolated common-sense reasoning, failing to capture reasoning ability across intricate physical dimensions. To address this, we introduce CSMBench, a dataset comprising 1,041 high-quality figures curated from premier journals up to September 2025. CSMBench categorizes data into four scientifically distinct regimes: atomic, micro, meso, and macro scales, strictly aligning with the focus and definitions in materials study. Through open-ended figure description and multiple-choice caption matching tasks, we evaluate state-of-the-art open-source and closed-source models. Our analysis identifies that performance varies significantly across physical scales due to the distinct visual characteristics, highlighting the limitations of current generalist models and identifying critical directions for achieving hierarchical and accurate understanding in materials research. The CSMBench is publicly released at: https://huggingface.co/datasets/lututu/CSMBench.

Xiujie Song, Qi Jia, Shota Watanabe et al.

An image can convey a compelling story by presenting rich, logically connected visual clues. These connections form Chains-of-Reasoning (CoRs) within the image, enabling viewers to infer events, causal relationships, and other information, thereby understanding the underlying story. In this paper, we focus on these semantically rich images and define them as Storytelling Images. Such images have diverse applications beyond illustration creation and cognitive screening, leveraging their ability to convey multi-layered information visually and inspire active interpretation. However, due to their complex semantic nature, Storytelling Images are inherently challenging to create, and thus remain relatively scarce. To address this challenge, we introduce the Storytelling Image Generation task, which explores how generative AI models can be leveraged to create such images. Specifically, we propose a two-stage pipeline, StorytellingPainter, which combines the creative reasoning abilities of Large Language Models (LLMs) with the visual synthesis capabilities of Text-to-Image (T2I) models to generate Storytelling Images. Alongside this pipeline, we develop a dedicated evaluation framework comprising three main evaluators: a Semantic Complexity Evaluator, a KNN-based Diversity Evaluator and a Story-Image Alignment Evaluator. Given the critical role of story generation in the Storytelling Image Generation task and the performance disparity between open-source and proprietary LLMs, we further explore tailored training strategies to reduce this gap, resulting in a series of lightweight yet effective models named Mini-Storytellers. Experimental results demonstrate the feasibility and effectiveness of our approaches. The code is available at https://github.com/xiujiesong/StorytellingImageGeneration.

Qi Jia, Yizhu Liu, Haifeng Tang et al.

Previous dialogue summarization techniques adapt large language models pretrained on the narrative text by injecting dialogue-specific features into the models. These features either require additional knowledge to recognize or make the resulting models harder to tune. To bridge the format gap between dialogues and narrative summaries in dialogue summarization tasks, we propose to post-train pretrained language models (PLMs) to rephrase from dialogue to narratives. After that, the model is fine-tuned for dialogue summarization as usual. Comprehensive experiments show that our approach significantly improves vanilla PLMs on dialogue summarization and outperforms other SOTA models by the summary quality and implementation costs.

Cunjuan Zhu, Qi Jia, Wei Chen et al.

Video-Text Retrieval (VTR) aims to search for the most relevant video related to the semantics in a given sentence, and vice versa. In general, this retrieval task is composed of four successive steps: video and textual feature representation extraction, feature embedding and matching, and objective functions. In the last, a list of samples retrieved from the dataset is ranked based on their matching similarities to the query. In recent years, significant and flourishing progress has been achieved by deep learning techniques, however, VTR is still a challenging task due to the problems like how to learn an efficient spatial-temporal video feature and how to narrow the cross-modal gap. In this survey, we review and summarize over 100 research papers related to VTR, demonstrate state-of-the-art performance on several commonly benchmarked datasets, and discuss potential challenges and directions, with the expectation to provide some insights for researchers in the field of video-text retrieval.

Qi Jia, Yizhu Liu, Haifeng Tang et al.

Curriculum learning has shown promising improvements in multiple domains by training machine learning models from easy samples to hard ones. Previous works which either design rules or train models for scoring the difficulty highly rely on task-specific expertise, and cannot generalize. Inspired by the "easy-to-hard" intuition, we propose to do in-sample curriculum learning for natural language generation tasks. Our learning strategy starts training the model to generate the last few words, i.e., do sequence completion, and gradually extends to generate the whole output sequence. Comprehensive experiments show that it generalizes well to different tasks and achieves significant improvements over strong baselines.

Xianfeng Wang, Kaiwei Zhang, Qi Jia et al.

While Multimodal Large Language Models (MLLMs) have demonstrated impressive proficiency in high-level reasoning tasks, such as complex diagrammatic interpretation, it remains an open question whether they possess the fundamental visual primitives comparable to human intuition. To investigate this, we introduce KidVis, a novel benchmark grounded in the theory of human visual development. KidVis deconstructs visual intelligence into six atomic capabilities - Concentration, Tracking, Discrimination, Memory, Spatial, and Closure - already possessed by 6-7 year old children, comprising 10 categories of low-semantic-dependent visual tasks. Evaluating 20 state-of-the-art MLLMs against a human physiological baseline reveals a stark performance disparity. Results indicate that while human children achieve a near-perfect average score of 95.32, the state-of-the-art GPT-5 attains only 67.33. Crucially, we observe a "Scaling Law Paradox": simply increasing model parameters fails to yield linear improvements in these foundational visual capabilities. This study confirms that current MLLMs, despite their reasoning prowess, lack the essential physiological perceptual primitives required for generalized visual intelligence.

Qi Jia, Siyu Ren, Yizhu Liu et al.

Despite tremendous improvements in natural language generation, summarization models still suffer from the unfaithfulness issue. Previous work evaluates faithfulness either using models trained on the other tasks or in-domain synthetic data, or prompting a large model such as ChatGPT. This paper proposes to do zero-shot faithfulness evaluation simply with a moderately-sized foundation language model. We introduce a new metric FFLM, which is a combination of probability changes based on the intuition that prefixing a piece of text that is consistent with the output will increase the probability of predicting the output. Experiments show that FFLM performs competitively with or even outperforms ChatGPT on both inconsistency detection and faithfulness rating with 24x fewer parameters. FFLM also achieves improvements over other strong baselines.

Xiangyang Zhu, Yuan Tian, Zicheng Zhang et al.

Large vision-language models (LVLMs) exhibit remarkable capabilities in cross-modal tasks but face significant safety challenges, which undermine their reliability in real-world applications. Efforts have been made to build LVLM safety evaluation benchmarks to uncover their vulnerability. However, existing benchmarks are hindered by their labor-intensive construction process, static complexity, and limited discriminative power. Thus, they may fail to keep pace with rapidly evolving models and emerging risks. To address these limitations, we propose VLSafetyBencher, the first automated system for LVLM safety benchmarking. VLSafetyBencher introduces four collaborative agents: Data Preprocessing, Generation, Augmentation, and Selection agents to construct and select high-quality samples. Experiments validates that VLSafetyBencher can construct high-quality safety benchmarks within one week at a minimal cost. The generated benchmark effectively distinguish safety, with a safety rate disparity of 70% between the most and least safe models.

Qi Jia, Xiang Yue, Tianyu Zheng et al.

We introduce SimulBench, a benchmark designed to evaluate large language models (LLMs) across a diverse collection of creative simulation scenarios, such as acting as a Linux terminal or playing text games with users. While these simulation tasks serve as effective measures of an LLM's general intelligence, they are seldom incorporated into existing benchmarks. A major challenge is to develop an evaluation framework for testing different LLMs fairly while preserving the multi-round interactive nature of simulation tasks between users and AI. To tackle this issue, we suggest using a fixed LLM as a user agent to engage with an LLM to collect dialogues first under different tasks. Then, challenging dialogue scripts are extracted for evaluating different target LLMs. To facilitate automatic assessment on \DataName{}, GPT-4 is employed as the evaluator, tasked with reviewing the quality of the final response generated by the target LLMs given multi-turn dialogue scripts. Our comprehensive experiments indicate that these simulation tasks continue to pose a significant challenge with their unique natures and show the gap between proprietary models and the most advanced open LLMs. For example, GPT-4-turbo outperforms LLaMA-3-70b-Chat on 18.55\% more cases.

Yu Liu, Yaqi Cai, Qi Jia et al.

Ultra-fine-grained visual categorization (Ultra-FGVC) aims at distinguishing highly similar sub-categories within fine-grained objects, such as different soybean cultivars. Compared to traditional fine-grained visual categorization, Ultra-FGVC encounters more hurdles due to the small inter-class and large intra-class variation. Given these challenges, relying on human annotation for Ultra-FGVC is impractical. To this end, our work introduces a novel task termed Ultra-Fine-Grained Novel Class Discovery (UFG-NCD), which leverages partially annotated data to identify new categories of unlabeled images for Ultra-FGVC. To tackle this problem, we devise a Region-Aligned Proxy Learning (RAPL) framework, which comprises a Channel-wise Region Alignment (CRA) module and a Semi-Supervised Proxy Learning (SemiPL) strategy. The CRA module is designed to extract and utilize discriminative features from local regions, facilitating knowledge transfer from labeled to unlabeled classes. Furthermore, SemiPL strengthens representation learning and knowledge transfer with proxy-guided supervised learning and proxy-guided contrastive learning. Such techniques leverage class distribution information in the embedding space, improving the mining of subtle differences between labeled and unlabeled ultra-fine-grained classes. Extensive experiments demonstrate that RAPL significantly outperforms baselines across various datasets, indicating its effectiveness in handling the challenges of UFG-NCD. Code is available at https://github.com/SSDUT-Caiyq/UFG-NCD.

Ping Guo, Chengzhou Li, Guanchen Meng et al.

As one of the most important underwater sensing technologies, forward-looking sonar exhibits unique imaging characteristics. Sonar images are often affected by severe speckle noise, low texture contrast, acoustic shadows, and geometric distortions. These factors make it difficult for traditional teacher-student frameworks to achieve satisfactory performance in sonar semantic segmentation tasks under extremely limited labeled data conditions. To address this issue, we propose a Collaborative Teacher Semantic Segmentation Framework for forward-looking sonar images. This framework introduces a multi-teacher collaborative mechanism composed of one general teacher and multiple sonar-specific teachers. By adopting a multi-teacher alternating guidance strategy, the student model can learn general semantic representations while simultaneously capturing the unique characteristics of sonar images, thereby achieving more comprehensive and robust feature modeling. Considering the challenges of sonar images, which can lead teachers to generate a large number of noisy pseudo-labels, we further design a cross-teacher reliability assessment mechanism. This mechanism dynamically quantifies the reliability of pseudo-labels by evaluating the consistency and stability of predictions across multiple views and multiple teachers, thereby mitigating the negative impact caused by noisy pseudo-labels. Notably, on the FLSMD dataset, when only 2% of the data is labeled, our method achieves a 5.08% improvement in mIoU compared to other state-of-the-art approaches.

Tianyu Zheng, Shuyue Guo, Xingwei Qu et al.

In this paper, we introduce Kun, a novel approach for creating high-quality instruction-tuning datasets for large language models (LLMs) without relying on manual annotations. Adapting a self-training algorithm based on instruction back-translation and answer polishment, Kun leverages unlabelled data from diverse sources such as Wudao, Wanjuan, and SkyPile to generate a substantial dataset of over a million Chinese instructional data points. This approach significantly deviates from traditional methods by using a self-curation process to refine and select the most effective instruction-output pairs. Our experiments with the 6B-parameter Yi model across various benchmarks demonstrate Kun's robustness and scalability. Our method's core contributions lie in its algorithmic advancement, which enhances data retention and clarity, and its innovative data generation approach that substantially reduces the reliance on costly and time-consuming manual annotations. This methodology presents a scalable and efficient solution for improving the instruction-following capabilities of LLMs, with significant implications for their application across diverse fields. The code and dataset can be found at https://github.com/Zheng0428/COIG-Kun

Xinwei Xue, Gaoyu Wang, Long Ma et al.

More and more attention has been paid to the segmentation of pulmonary nodules. Among the current methods based on deep learning, 3D segmentation methods directly input 3D images, which takes up a lot of memory and brings huge computation. However, most of the 2D segmentation methods with less parameters and calculation have the problem of lacking spatial relations between slices, resulting in poor segmentation performance. In order to solve these problems, we propose an adjacent slice feature guided 2.5D network. In this paper, we design an adjacent slice feature fusion model to introduce information from adjacent slices. To further improve the model performance, we construct a multi-scale fusion module to capture more context information, in addition, we design an edge-constrained loss function to optimize the segmentation results in the edge region. Fully experiments show that our method performs better than other existing methods in pulmonary nodule segmentation task.

Runze Zhang, Guoguang Du, Xiaochuan Li et al.

Spatio-temporal consistency is a critical research topic in video generation. A qualified generated video segment must ensure plot plausibility and coherence while maintaining visual consistency of objects and scenes across varying viewpoints. Prior research, especially in open-source projects, primarily focuses on either temporal or spatial consistency, or their basic combination, such as appending a description of a camera movement after a prompt without constraining the outcomes of this movement. However, camera movement may introduce new objects to the scene or eliminate existing ones, thereby overlaying and affecting the preceding narrative. Especially in videos with numerous camera movements, the interplay between multiple plots becomes increasingly complex. This paper introduces and examines integral spatio-temporal consistency, considering the synergy between plot progression and camera techniques, and the long-term impact of prior content on subsequent generation. Our research encompasses dataset construction through to the development of the model. Initially, we constructed a DropletVideo-10M dataset, which comprises 10 million videos featuring dynamic camera motion and object actions. Each video is annotated with an average caption of 206 words, detailing various camera movements and plot developments. Following this, we developed and trained the DropletVideo model, which excels in preserving spatio-temporal coherence during video generation. The DropletVideo dataset and model are accessible at https://dropletx.github.io.

Xiangyang Zhu, Yuan Tian, Qi Jia et al.

The success of large language models (LLMs) in scientific domains has heightened safety concerns, prompting numerous benchmarks to evaluate their scientific safety. Existing benchmarks often suffer from limited risk coverage and a reliance on subjective evaluation. To address these problems, we introduce SafeSci, a comprehensive framework for safety evaluation and enhancement in scientific contexts. SafeSci comprises SafeSciBench, a multi-disciplinary benchmark with 0.25M samples, and SafeSciTrain, a large-scale dataset containing 1.5M samples for safety enhancement. SafeSciBench distinguishes between safety knowledge and risk to cover extensive scopes and employs objective metrics such as deterministically answerable questions to mitigate evaluation bias. We evaluate 24 advanced LLMs, revealing critical vulnerabilities in current models. We also observe that LLMs exhibit varying degrees of excessive refusal behaviors on safety-related issues. For safety enhancement, we demonstrate that fine-tuning on SafeSciTrain significantly enhances the safety alignment of models. Finally, we argue that knowledge is a double-edged sword, and determining the safety of a scientific question should depend on specific context, rather than universally categorizing it as safe or unsafe. Our work provides both a diagnostic tool and a practical resource for building safer scientific AI systems.

Kaiwei Zhang, Qi Jia, Zijian Chen et al.

Large language models (LLMs), while increasingly used in domains requiring factual rigor, often display a troubling behavior: sycophancy, the tendency to align with user beliefs regardless of correctness. This tendency is reinforced by preference-based alignment techniques that optimize for user satisfaction but can undermine truthfulness. While relatively benign in casual dialogue, sycophancy poses serious risks in high-stakes settings such as scientific question answering (QA), where model outputs may shape collaborative reasoning, decision-making, and knowledge formation. Despite its importance, this phenomenon remains underexamined in factual QA contexts. We address this gap by introducing a unified evaluation framework to quantify the impact of sycophantic context on model behavior in scientific QA, measuring how much user-imposed social pressure distorts model outputs. The framework incorporates adversarial prompting setups and targeted metrics, such as misleading resistance and sycophancy resistance, that capture a model's ability to maintain factual consistency under misleading cues. Systematic evaluations across open-source and proprietary models reveal pervasive sycophantic tendencies, driven more by alignment strategy than by model size. To mitigate this issue, we propose Pressure-Tune, a lightweight post-training method that fine-tunes models on synthetic adversarial dialogues paired with chain-of-thought rationales. These rationales reject user misinformation while reinforcing factual commitments. Experiments on challenging scientific QA benchmarks show that Pressure-Tune significantly enhances sycophancy resistance without compromising accuracy or responsiveness to valid feedback, offering a practical pathway toward more truthful and principled model behavior.

Zijian Chen, Lirong Deng, Zhengyu Chen et al.

Evaluating the abilities of large models and manifesting their gaps are challenging. Current benchmarks adopt either ground-truth-based score-form evaluation on static datasets or indistinct textual chatbot-style human preferences collection, which may not provide users with immediate, intuitive, and perceptible feedback on performance differences. In this paper, we introduce BioMotion Arena, a novel framework for evaluating large language models (LLMs) and multimodal large language models (MLLMs) via visual animation. Our methodology draws inspiration from the inherent visual perception of motion patterns characteristic of living organisms that utilizes point-light source imaging to amplify the performance discrepancies between models. Specifically, we employ a pairwise comparison evaluation and collect more than 45k votes for 53 mainstream LLMs and MLLMs on 90 biological motion variants. Data analyses show that the crowd-sourced human votes are in good agreement with those of expert raters, demonstrating the superiority of our BioMotion Arena in offering discriminative feedback. We also find that over 90\% of evaluated models, including the cutting-edge open-source InternVL3 and proprietary Claude-4 series, fail to produce fundamental humanoid point-light groups, much less smooth and biologically plausible motions. This enables BioMotion Arena to serve as a challenging benchmark for performance visualization and a flexible evaluation framework without restrictions on ground-truth.

Junying Wang, Zicheng Zhang, Yijin Guo et al.

As foundation models grow rapidly in capability and deployment, evaluating their scientific understanding becomes increasingly critical. Existing science benchmarks have made progress towards broad Range, wide Reach, and high Rigor, yet they often face two major challenges: data leakage risks that compromise benchmarking validity, and evaluation inefficiency due to large-scale testing. To address these issues, we introduce the Ever-Evolving Science Exam (EESE), a dynamic benchmark designed to reliably assess scientific capabilities in foundation models. Our approach consists of two components: 1) a non-public EESE-Pool with over 100K expertly constructed science instances (question-answer pairs) across 5 disciplines and 500+ subfields, built through a multi-stage pipeline ensuring Range, Reach, and Rigor, 2) a periodically updated 500-instance subset EESE, sampled and validated to enable leakage-resilient, low-overhead evaluations. Experiments on 32 open- and closed-source models demonstrate that EESE effectively differentiates the strengths and weaknesses of models in scientific fields and cognitive dimensions. Overall, EESE provides a robust, scalable, and forward-compatible solution for science benchmark design, offering a realistic measure of how well foundation models handle science questions. The project page is at: https://github.com/aiben-ch/EESE.

Qi Jia, Kaiwei Zhang, Xiujie Song et al.

Understanding how well large language models can follow users' instructions throughout a dialogue spanning multiple topics is of great importance for data-intensive conversational applications. Existing benchmarks are often limited to a fixed number of turns, making them susceptible to saturation and failing to account for the user's interactive experience. In this work, we propose an extensible framework for assessing multi-turn instruction-following ability. At its core, our framework decouples linguistic surface forms from user intent simulation through a three-layer mechanism that tracks constraints, instructions, and topics. This framework mimics User-LLM interaction by enabling the dynamic construction of benchmarks with state changes and tracebacks, terminating a conversation only when the model exhausts a simulated user's patience. We define a suite of metrics capturing the quality of the interaction process. Using this framework, we construct EvolIF, an evolving instruction-following benchmark incorporating nine distinct constraint types. Our results indicate that GPT-5 exhibits superior instruction-following performance. It sustains an average of 18.54 conversational turns and demonstrates 70.31% robustness, outperforming Gemini-2.5-Pro by a significant margin of 11.41%, while other models lag far behind. All of the data and code will be made publicly available online.

Qi Jia, Siyu Ren, Ziheng Qin et al.

Datasets nowadays are generally constructed from multiple sources and using different synthetic techniques, making data de-noising and de-duplication crucial before being used for post-training. In this work, we propose to perform instruction tuning by iterative data selection (\ApproachName{}). We measure the quality of a sample from complexity and diversity simultaneously. Instead of calculating the complexity score once for all before fine-tuning, we highlight the importance of updating this model-specific score during fine-tuning to accurately accommodate the dynamic changes of the model. On the other hand, the diversity score is defined on top of the samples' responses under the consideration of their informativeness. IterIT integrates the strengths of both worlds by iteratively updating the complexity score for the top-ranked samples and greedily selecting the ones with the highest complexity-diversity score. Experiments on multiple instruction-tuning data demonstrate consistent improvements of IterIT over strong baselines. Moreover, our approach also generalizes well to domain-specific scenarios and different backbone models. All resources will be available at https://github.com/JiaQiSJTU/IterIT.

Xiaochuan Li, Guoguang Du, Runze Zhang et al.

Scaling laws have validated the success and promise of large-data-trained models in creative generation across text, image, and video domains. However, this paradigm faces data scarcity in the 3D domain, as there is far less of it available on the internet compared to the aforementioned modalities. Fortunately, there exist adequate videos that inherently contain commonsense priors, offering an alternative supervisory signal to mitigate the generalization bottleneck caused by limited native 3D data. On the one hand, videos capturing multiple views of an object or scene provide a spatial consistency prior for 3D generation. On the other hand, the rich semantic information contained within the videos enables the generated content to be more faithful to the text prompts and semantically plausible. This paper explores how to apply the video modality in 3D asset generation, spanning datasets to models. We introduce Droplet3D-4M, the first large-scale video dataset with multi-view level annotations, and train Droplet3D, a generative model supporting both image and dense text input. Extensive experiments validate the effectiveness of our approach, demonstrating its ability to produce spatially consistent and semantically plausible content. Moreover, in contrast to the prevailing 3D solutions, our approach exhibits the potential for extension to scene-level applications. This indicates that the commonsense priors from the videos significantly facilitate 3D creation. We have open-sourced all resources including the dataset, code, technical framework, and model weights: https://dropletx.github.io/.

Ziheng Qin, Hailun Xu, Wei Chee Yew et al.

Machine learning relies heavily on data, yet the continuous growth of real-world data poses challenges for efficient dataset construction and training. A fundamental yet unsolved question is: given our current model and data, does a new data (sample/batch) need annotation/learning? Conventional approaches retain all available data, leading to non-optimal data and training efficiency. Active learning aims to reduce data redundancy by selecting a subset of samples to annotate, while it increases pipeline complexity and introduces bias. In this work, we propose Info-Coevolution, a novel framework that efficiently enables models and data to coevolve through online selective annotation with no bias. Leveraging task-specific models (and open-source models), it selectively annotates and integrates online and web data to improve datasets efficiently. For real-world datasets like ImageNet-1K, Info-Coevolution reduces annotation and training costs by 32\% without performance loss. It is able to automatically give the saving ratio without tuning the ratio. It can further reduce the annotation ratio to 50\% with semi-supervised learning. We also explore retrieval-based dataset enhancement using unlabeled open-source data. Code is available at https://github.com/NUS-HPC-AI-Lab/Info-Coevolution/.

Fei Du, Peng Yang, Qi Jia et al.

In this paper, our goal is to design a simple learning paradigm for long-tail visual recognition, which not only improves the robustness of the feature extractor but also alleviates the bias of the classifier towards head classes while reducing the training skills and overhead. We propose an efficient one-stage training strategy for long-tailed visual recognition called Global and Local Mixture Consistency cumulative learning (GLMC). Our core ideas are twofold: (1) a global and local mixture consistency loss improves the robustness of the feature extractor. Specifically, we generate two augmented batches by the global MixUp and local CutMix from the same batch data, respectively, and then use cosine similarity to minimize the difference. (2) A cumulative head tail soft label reweighted loss mitigates the head class bias problem. We use empirical class frequencies to reweight the mixed label of the head-tail class for long-tailed data and then balance the conventional loss and the rebalanced loss with a coefficient accumulated by epochs. Our approach achieves state-of-the-art accuracy on CIFAR10-LT, CIFAR100-LT, and ImageNet-LT datasets. Additional experiments on balanced ImageNet and CIFAR demonstrate that GLMC can significantly improve the generalization of backbones. Code is made publicly available at https://github.com/ynu-yangpeng/GLMC.

Qi Jia, Haodong Zhao, Dun Pei et al.

Benchmarking AI systems in multi-turn interactive scenarios is essential for understanding their practical capabilities in real-world applications. However, existing evaluation protocols are highly heterogeneous, differing significantly in dataset formats, model interfaces, and evaluation pipelines, which severely impedes systematic comparison. In this work, we present UniDial-EvalKit (UDE), a unified evaluation toolkit for assessing interactive AI systems. The core contribution of UDE lies in its holistic unification: it standardizes heterogeneous data formats into a universal schema, streamlines complex evaluation pipelines through a modular architecture, and aligns metric calculations under a consistent scoring interface. It also supports efficient large-scale evaluation through parallel generation and scoring, as well as checkpoint-based caching to eliminate redundant computation. Validated across diverse multi-turn benchmarks, UDE not only guarantees high reproducibility through standardized workflows and transparent logging, but also significantly improves evaluation efficiency and extensibility. We make the complete toolkit and evaluation scripts publicly available to foster a standardized benchmarking ecosystem and accelerate future breakthroughs in interactive AI.

Marcos V. Conde, Zhijun Lei, Wen Li et al.

This paper introduces a novel benchmark as part of the AIS 2024 Real-Time Image Super-Resolution (RTSR) Challenge, which aims to upscale compressed images from 540p to 4K resolution (4x factor) in real-time on commercial GPUs. For this, we use a diverse test set containing a variety of 4K images ranging from digital art to gaming and photography. The images are compressed using the modern AVIF codec, instead of JPEG. All the proposed methods improve PSNR fidelity over Lanczos interpolation, and process images under 10ms. Out of the 160 participants, 25 teams submitted their code and models. The solutions present novel designs tailored for memory-efficiency and runtime on edge devices. This survey describes the best solutions for real-time SR of compressed high-resolution images.

Yanyi Zhang, Qi Jia, Xin Fan et al.

Attribute and object (A-O) disentanglement is a fundamental and critical problem for Compositional Zero-shot Learning (CZSL), whose aim is to recognize novel A-O compositions based on foregone knowledge. Existing methods based on disentangled representation learning lose sight of the contextual dependency between the A-O primitive pairs. Inspired by this, we propose a novel A-O disentangled framework for CZSL, namely Class-specified Cascaded Network (CSCNet). The key insight is to firstly classify one primitive and then specifies the predicted class as a priori for guiding another primitive recognition in a cascaded fashion. To this end, CSCNet constructs Attribute-to-Object and Object-to-Attribute cascaded branches, in addition to a composition branch modeling the two primitives as a whole. Notably, we devise a parametric classifier (ParamCls) to improve the matching between visual and semantic embeddings. By improving the A-O disentanglement, our framework achieves superior results than previous competitive methods.

Farong Wen, Yijin Guo, Junying Wang et al.

The rapid development of Multimodal Large Language Models (MLLM) has led to a wide range of MLLM applications, and a number of benchmark datasets have sprung up in order to assess MLLM abilities. However, full-coverage Q&A testing on large-scale data is resource-intensive and time-consuming. To address this issue, we propose the MLLM Interview (MITV) strategy, which aims to quickly obtain MLLM performance metrics by quizzing fewer question. First, First, we constructed the interview dataset, which was built on an existing MLLM assessment dataset, by adding difficulty labels based on the performance of some typical MLLMs in this dataset. Second, we propose an MLLM Interview strategy, which obtains an initial performance situation of the large model by quizzing a small number of topics and then continuously tries to test the model's limits. Through extensive experiments, the result shows that the MITV strategy proposed in this paper performs well on MLLM benchmark datasets, and it is able to obtain the model evaluation capability faster through a small number of questions and answers.

Qi Jia, Xiujie Song, Zicheng Zhang et al.

Existing benchmarks for large language models (LLMs) predominantely focus on assessing their capabilities through verifiable tasks. Such objective and static benchmarks offer limited utility for practical LLM selection, making it difficult for users to find suitable models for their individual needs. To bridge this gap, we present the first User-Centric Subjective Leaderboard (USL), which provides a preference-driven, dynamic ranking of LLMs across diverse real-world scenarios. Our work is built upon a thorough investigation of real human preference data, involving more than 10K subjective queries. Our investigation reveals significant diversity and contradictions in human preferences, which limit the effectiveness of state-of-the-art reward models. To address this, we introduce Customizable Reward Models (CRMs). With only 4B parameters, our CRM surpasses the performance of leading models such as GPT-4.1 and Gemini-2.5-pro, showing exceptional generalization capabilities across new topics and criteria. The USL, powered by CRMs, exhibits strong negative correlations to contradictory preferences.

Chengzhou Li, Ping Guo, Guanchen Meng et al.

Object detection in sonar images is a key technology in underwater detection systems. Compared to natural images, sonar images contain fewer texture details and are more susceptible to noise, making it difficult for non-experts to distinguish subtle differences between classes. This leads to their inability to provide precise annotation data for sonar images. Therefore, designing effective object detection methods for sonar images with extremely limited labels is particularly important. To address this, we propose a teacher-student framework called RSOD, which aims to fully learn the characteristics of sonar images and develop a pseudo-label strategy suitable for these images to mitigate the impact of limited labels. First, RSOD calculates a reliability score by assessing the consistency of the teacher's predictions across different views. To leverage this score, we introduce an object mixed pseudo-label method to tackle the shortage of labeled data in sonar images. Finally, we optimize the performance of the student by implementing a reliability-guided adaptive constraint. By taking full advantage of unlabeled data, the student can perform well even in situations with extremely limited labels. Notably, on the UATD dataset, our method, using only 5% of labeled data, achieves results that can compete against those of our baseline algorithm trained on 100% labeled data. We also collected a new dataset to provide more valuable data for research in the field of sonar.

Yunhao Li, Sijing Wu, Huiyu Duan et al.

In recent years, with the rapid development of powerful multimodal large language models (MLLMs), explainable image quality assessment (IQA) has gradually become popular, aiming at providing quality-related descriptions and answers of images. To achieve this goal, recent methods seek to construct a large-scale instruction tuning dataset to empower the MLLM with quality perception ability following the well-known scaling law. However, a large amount of instruction tuning data may cause substantial computational costs and redundant data, which in turn will cause harm to the performance of the model. To cope with this problem, in this paper, we challenge the scaling law and systematically investigate the role of data quality of the instruction tuning dataset for explainable IQA. Using a powerful pre-trained MLLM, we first investigate the changes in model performance after fine-tuning with different sizes of instruction tuning data. We find that selecting a subset of the data set randomly using an appropriate ratio can even lead to better results than training with the entire instruction tuning dataset, demonstrating the redundancy of current explainable IQA instruction tuning data. Beyond randomly sampling a subset, we propose a clustering-based data selection framework with three stages: clustering feature extraction, cluster quota allocation, and cluster sampling strategy. Then we systematically analyze the choices of each stage and propose a simple but efficient data selection method IQA-Select for explainable IQA. The experimental results demonstrate that IQA-Select can achieve 102.1% and 103.7% performance of full fine-tuning using only 10% selected data in Q-Bench and AesBench respectively, significantly reducing computational costs while achieving better performance.

Junying Wang, Zicheng Zhang, Ye Shen et al.

High-quality, multi-modal benchmarks are crucial for advancing scientific reasoning in large models yet their manual creation is costly and unscalable. To address this bottleneck, we explore the potential for transforming Text-Only QA Pairs (TQAs) into high-quality Multi-Modal QA Pairs (MMQAs), which include three parts: 1) Task Definition \& Evaluation Rubric: We develop a TQA-to-MMQA framework and establish a comprehensive, multi-dimensional MMQA quality rubric that provides principles for the transformation. 2) Benchmark Construction: Then we construct two extensive benchmarks to rigorously evaluate state-of-the-art generation \& understanding models on the distinct tasks of MMQA generation \& MMQA quality evaluation. 3) Preliminary Solution: We develop an agentic system (Q-Mirror), which operationalizes our framework by integrating MMQA generation and evaluation into a closed loop for iterative refinement. Our experiments show that while state-of-the-art models can generate MMQAs, their outputs still leave substantial gaps, underscoring the need for reliable evaluation. We further demonstrate that top-tier understanding models align closely with human judgment in MMQA quality assessment. Leveraging both insights, the Q-Mirror agent raises average scores from 78.90 to 85.22 and pass rates from 72\% to 95\%, offering a practical path to large-scale scientific benchmarks.

Yijin Guo, Zicheng Zhang, Ye Shen et al.

Existing evaluations of foundation models, including recent human-centric approaches, fail to capture what truly matters: user's experience during interaction. Current methods treat evaluation as a matter of output correctness alone, overlooking that user satisfaction emerges from the interplay between response quality and interaction, which limits their ability to account for the mechanisms underlying user experience. To address this gap, we introduce QoNext, the first framework that adapts Quality of Experience (QoE) principles from networking and multimedia to the assessment of foundation models. QoNext identifies experiential factors that shape user experience and incorporates them into controlled experiments, where human ratings are collected under varied configurations. From these studies we construct a QoE-oriented database and train predictive models that estimate perceived user experience from measurable system parameters. Our results demonstrate that QoNext not only enables proactive and fine-grained evaluation but also provides actionable guidance for productized services of optimizing foundation models in practice.

Ye Shen, Junying Wang, Farong Wen et al.

The rapid progress of Multi-Modal Large Language Models (MLLMs) has spurred the creation of numerous benchmarks. However, conventional full-coverage Question-Answering evaluations suffer from high redundancy and low efficiency. Inspired by human interview processes, we propose a multi-to-one interview paradigm for efficient MLLM evaluation. Our framework consists of (i) a two-stage interview strategy with pre-interview and formal interview phases, (ii) dynamic adjustment of interviewer weights to ensure fairness, and (iii) an adaptive mechanism for question difficulty-level chosen. Experiments on different benchmarks show that the proposed paradigm achieves significantly higher correlation with full-coverage results than random sampling, with improvements of up to 17.6% in PLCC and 16.7% in SRCC, while reducing the number of required questions. These findings demonstrate that the proposed paradigm provides a reliable and efficient alternative for large-scale MLLM benchmarking.

Yihua Wang, Qi Jia, Cong Xu et al.

Multimodal sarcasm detection is a complex task that requires distinguishing subtle complementary signals across modalities while filtering out irrelevant information. Many advanced methods rely on learning shortcuts from datasets rather than extracting intended sarcasm-related features. However, our experiments show that shortcut learning impairs the model's generalization in real-world scenarios. Furthermore, we reveal the weaknesses of current modality fusion strategies for multimodal sarcasm detection through systematic experiments, highlighting the necessity of focusing on effective modality fusion for complex emotion recognition. To address these challenges, we construct MUStARD++$^{R}$ by removing shortcut signals from MUStARD++. Then, a Multimodal Conditional Information Bottleneck (MCIB) model is introduced to enable efficient multimodal fusion for sarcasm detection. Experimental results show that the MCIB achieves the best performance without relying on shortcut learning.

Yanyi Zhang, Binglin Qiu, Qi Jia et al.

Most incremental learners excessively prioritize coarse classes of objects while neglecting various kinds of states (e.g. color and material) attached to the objects. As a result, they are limited in the ability to reason fine-grained compositionality of state-object pairs. To remedy this limitation, we propose a novel task called Compositional Incremental Learning (composition-IL), enabling the model to recognize state-object compositions as a whole in an incremental learning fashion. Since the lack of suitable benchmarks, we re-organize two existing datasets and make them tailored for composition-IL. Then, we propose a prompt-based Composition Incremental Learner (CompILer), to overcome the ambiguous composition boundary problem which challenges composition-IL largely. Specifically, we exploit multi-pool prompt learning, which is regularized by inter-pool prompt discrepancy and intra-pool prompt diversity. Besides, we devise object-injected state prompting by using object prompts to guide the selection of state prompts. Furthermore, we fuse the selected prompts by a generalized-mean strategy, to eliminate irrelevant information learned in the prompts. Extensive experiments on two datasets exhibit state-of-the-art performance achieved by CompILer.

Qi Jia, Baoyu Fan, Cong Xu et al.

Existing video multi-modal sentiment analysis mainly focuses on the sentiment expression of people within the video, yet often neglects the induced sentiment of viewers while watching the videos. Induced sentiment of viewers is essential for inferring the public response to videos, has broad application in analyzing public societal sentiment, effectiveness of advertising and other areas. The micro videos and the related comments provide a rich application scenario for viewers induced sentiment analysis. In light of this, we introduces a novel research task, Multi-modal Sentiment Analysis for Comment Response of Video Induced(MSA-CRVI), aims to inferring opinions and emotions according to the comments response to micro video. Meanwhile, we manually annotate a dataset named Comment Sentiment toward to Micro Video (CSMV) to support this research. It is the largest video multi-modal sentiment dataset in terms of scale and video duration to our knowledge, containing 107,267 comments and 8,210 micro videos with a video duration of 68.83 hours. To infer the induced sentiment of comment should leverage the video content, so we propose the Video Content-aware Comment Sentiment Analysis (VC-CSA) method as baseline to address the challenges inherent in this new task. Extensive experiments demonstrate that our method is showing significant improvements over other established baselines.

Qi Jia, Haifeng Tang, Kenny Q. Zhu

Changing speaker names consistently throughout a dialogue should not affect its meaning and corresponding outputs for text generation from dialogues. However, pre-trained language models, serving as the backbone for dialogue-processing tasks, have shown to be sensitive to nuances. This may result in unfairness in real-world applications. No comprehensive analysis of this problem has been done in the past. In this work, we propose to quantitatively measure a model's sensitivity on speaker names, and comprehensively evaluate a number of known methods for reducing speaker name sensitivity, including a novel approach of our own. Extensive experiments on multiple datasets provide a benchmark for this problem and show the favorable performance of our approach in sensitivity reduction and quality of generation.

Yuqing Liu, Qi Jia, Xin Fan et al.

It is challenging to restore low-resolution (LR) images to super-resolution (SR) images with correct and clear details. Existing deep learning works almost neglect the inherent structural information of images, which acts as an important role for visual perception of SR results. In this paper, we design a hierarchical feature exploitation network to probe and preserve structural information in a multi-scale feature fusion manner. First, we propose a cross convolution upon traditional edge detectors to localize and represent edge features. Then, cross convolution blocks (CCBs) are designed with feature normalization and channel attention to consider the inherent correlations of features. Finally, we leverage multi-scale feature fusion group (MFFG) to embed the cross convolution blocks and develop the relations of structural features in different scales hierarchically, invoking a lightweight structure-preserving network named as Cross-SRN. Experimental results demonstrate the Cross-SRN achieves competitive or superior restoration performances against the state-of-the-art methods with accurate and clear structural details. Moreover, we set a criterion to select images with rich structural textures. The proposed Cross-SRN outperforms the state-of-the-art methods on the selected benchmark, which demonstrates that our network has a significant advantage in preserving edges.

Qi Jia, Hongru Huang, Kenny Q. Zhu

Interpersonal language style shifting in dialogues is an interesting and almost instinctive ability of human. Understanding interpersonal relationship from language content is also a crucial step toward further understanding dialogues. Previous work mainly focuses on relation extraction between named entities in texts. In this paper, we propose the task of relation classification of interlocutors based on their dialogues. We crawled movie scripts from IMSDb, and annotated the relation labels for each session according to 13 pre-defined relationships. The annotated dataset DDRel consists of 6300 dyadic dialogue sessions between 694 pair of speakers with 53,126 utterances in total. We also construct session-level and pair-level relation classification tasks with widely-accepted baselines. The experimental results show that this task is challenging for existing models and the dataset will be useful for future research.

Qi Jia, Yizhu Liu, Siyu Ren et al.

Multi-turn response selection is a task designed for developing dialogue agents. The performance on this task has a remarkable improvement with pre-trained language models. However, these models simply concatenate the turns in dialogue history as the input and largely ignore the dependencies between the turns. In this paper, we propose a dialogue extraction algorithm to transform a dialogue history into threads based on their dependency relations. Each thread can be regarded as a self-contained sub-dialogue. We also propose Thread-Encoder model to encode threads and candidates into compact representations by pre-trained Transformers and finally get the matching score through an attention layer. The experiments show that dependency relations are helpful for dialogue context understanding, and our model outperforms the state-of-the-art baselines on both DSTC7 and DSTC8*, with competitive results on UbuntuV2.