27 accepted papers
While real-world applications increasingly demand intricate scene manipulation, existing instruction-guided image editing benchmarks often oversimplify task complexity and lack comprehensive, fine-grained instructions. To bridge this gap, we introduce CompBench, a large-scale benchmark specifically
Determining an effective data mixture is a key factor in Large Language Model (LLM) pre-training, where models must balance general competence with proficiency on hard tasks such as math and code. However, identifying an optimal mixture remains an open challenge, as existing approaches either rely o…
Geometric shapes play important roles in both physical world and human cognition. While multimodal large language models (MLLMs) have made significant advancements in visual understanding, their abilities to recognize geometric shapes and their spatial relationships, which we term geometric percepti…
DeepSeek-R1-Zero has successfully demonstrated the emergence of reasoning capabilities in LLMs purely through Reinforcement Learning (RL). Inspired by this breakthrough, we explore how RL can be utilized to enhance the reasoning capability of MLLMs. However, direct training with RL struggles to act…
Acoustic Echo Cancellation (AEC) is an essential speech signal processing technology that removes echoes from microphone inputs to facilitate natural-sounding full-duplex communication. Currently, deep learning-based AEC methods primarily focus on refining model architectures, frequently neglecting…
Leveraging the generative power of diffusion models, generative image compression has achieved impressive perceptual fidelity even at extremely low bitrates. However, current methods often neglect the non-uniform complexity of images, limiting their ability to balance global perceptual quality with…
Multimodal Large Language Models (MLLMs) have achieved remarkable success in vision understanding, reasoning, and interaction. However, the inference computation and memory increase progressively with the generation of output tokens during decoding, directly affecting the efficacy of MLLMs. Existing…
While various multimodal multi-image evaluation datasets have been emerged, but these datasets are primarily based on English, and there has yet to be a Chinese multi-image dataset. To fill this gap, we introduce RealBench, the first Chinese multimodal multi-image dataset, which contains 9393 sample
With the rapid advancement of Social Networking Services (SNS), the need for intelligent and efficient interaction within diverse platforms has become more crucial. Large Language Models (LLMs) play an important role in SNS as they possess the potential to revolutionize user experience, content gene…
In recent years, user generated content (UGC) has become the dominant force in internet traffic. However, UGC videos exhibit a higher degree of variability and diverse characteristics compared to traditional encoding test videos. This variance challenges the effectiveness of data-driven machine lear…
Relation triplet extraction is a fundamental task in natural language processing that aims to identify semantic relationships between entities in text. It is particularly challenging in the zero-shot setting, i.e., zero-shot relation triplet extraction (ZeroRTE), where the relation sets between trai…
This work presents an ergonomic and interactive human-robot collaboration (HRC) framework, through which new collaborative skills are extracted from a one-shot human demonstration and learned through Riemannian dynamic movement primitives (DMP). The proposed framework responds to human-robot interac
Multimodal large language models (MLLMs) have attracted increasing attention in the past few years, but they may still generate descriptions that include objects not present in the corresponding images, a phenomenon known as object hallucination. To eliminate hallucinations, existing methods manuall…
There is still a lack of null-space impedance control defined in Cartesian space that is suitable for multipoint contact and does not require knowledge of the force locations. To address this problem, this letter demonstrates a type of Cartesian null-space impedance control for the anthropomorphic m
Text2SQL is a task that translates natural language into SQL statements. Context-dependent Text2SQL offers a more natural database interaction by simulating dialogues between users and databases, with CoSQL and SparC as representative datasets. Yet, these datasets struggle to accurately replicate re…
Since the difference in angular velocity is used as the derivative of the orientation error in the classical impedance control, there is no longer a form of the second-order differential equation (SODE), and there is non-linearity in the classical impedance control, which limits applications. To add
Safe robot motion generation is critical for practical applications from manufacturing to homes. In this work, we proposed a stochastic optimization-based motion generation method to generate collision-free and time-optimal motion for the articulated robot represented by composite signed distance fi
Multimodal Aspect-based Sentiment Analysis (MABSA) is a fine-grained Sentiment Analysis task, which has attracted growing research interests recently. Existing work mainly utilizes image information to improve the performance of MABSA task. However, most of the studies overestimate the importance of…
Aspect Sentiment Triplet Extraction (ASTE) is widely used in various applications. However, existing ASTE datasets are limited in their ability to represent real-world scenarios, hindering the advancement of research in this area. In this paper, we introduce a new dataset, named DMASTE, which is man…
Multi-Label Few-Shot Aspect Category Detection (FS-ACD) is a new sub-task of aspect-based sentiment analysis, which aims to detect aspect categories accurately with limited training instances. Recently, dominant works use the prototypical network to accomplish this task, and employ the attention mec…
Target-oriented multimodal sentiment classification (TMSC) is a new subtask of aspect-based sentiment analysis, which aims to determine the sentiment polarity of the opinion target mentioned in a (sentence, image) pair. Recently, dominant works employ the attention mechanism to capture the correspon…
In this work, we propose a framework to address the autonomous impedance regulation problem of robots in a class of constrained manipulation tasks. In this framework, a human arm endpoint stiffness model is used to extract the task stiffness geometry along the constrained trajectory, which is then e
This work presents a unified approach for hybrid motion control of the Mobile Collaborative Robotic Assistant (MOCA). The objective is to develop a loco-manipulation controller, enabling various couplings of the arm and the mobile base movements, and particularly their purely decoupled motions. The
Data augmentation is commonly used to help build a robust speaker verification system, especially in limited-resource case. However, conventional data augmentation methods usually focus on the diversity of acoustic environment, leaving the lexicon variation neglected. For text dependent speaker veri…
Aspect-level sentiment classification (ASC) aims to detect the sentiment polarity of a given opinion target in a sentence. In neural network-based methods for ASC, most works employ the attention mechanism to capture the corresponding sentiment words of the opinion target, then aggregate them as evi…
Recently, deep neural networks (DNNs) have achieved incredible performance in speaker verification. However, most of which remains sensitive to environment noise. In this paper, we propose an end-to-end speaker verification framework to enhance the robustness against background noise. The proposed f…
This letter presents a novel teleoperation interface that enables remote loco-manipulation control of a MObile Collaborative robotic Assistant (MOCA). MOCA is a new research platform developed at the Istituto Italiano di Tecnologia (IIT), which is composed of a lightweight manipulator arm, a Pisa/II