arXiv subjects

Sheng Li

Publications and source records attributed to Sheng Li.

At least 91 records · Page 5Linked to original sources

arXiv2025

Towards Emotion Co-regulation with LLM-powered Socially Assistive Robots: Integrating LLM Prompts and Robotic Behaviors to Support Parent-Neurodivergent Child Dyads

Socially Assistive Robotics (SAR) has shown promise in supporting emotion regulation for neurodivergent children. Recently, there has been increasing interest in leveraging advanced technologies to assist parents in co-regulating emotions with their children. However, limited research has explored the integration of large language models (LLMs) with SAR to facilitate emotion co-regulation between parents and children with neurodevelopmental disorders. To address this gap, we developed an LLM-powered social robot by deploying a speech communication module on the MiRo-E robotic platform. This supervised autonomous system integrates LLM prompts and robotic behaviors to deliver tailored interventions for both parents and neurodivergent children. Pilot tests were conducted with two parent-child dyads, followed by a qualitative analysis. The findings reveal MiRo-E's positive impacts on interaction dynamics and its potential to facilitate emotion regulation, along with identified design and technical challenges. Based on these insights, we provide design implications to advance the future development of LLM-powered SAR for mental health applications.

cs.HC↗

arXiv2025

End-to-end Acoustic-linguistic Emotion and Intent Recognition Enhanced by Semi-supervised Learning

Emotion and intent recognition from speech is essential and has been widely investigated in human-computer interaction. The rapid development of social media platforms, chatbots, and other technologies has led to a large volume of speech data streaming from users. Nevertheless, annotating such data manually is expensive, making it challenging to train machine learning models for recognition purposes. To this end, we propose applying semi-supervised learning to incorporate a large scale of unlabelled data alongside a relatively smaller set of labelled data. We train end-to-end acoustic and linguistic models, each employing multi-task learning for emotion and intent recognition. Two semi-supervised learning approaches, including fix-match learning and full-match learning, are compared. The experimental results demonstrate that the semi-supervised learning approaches improve model performance in speech emotion and intent recognition from both acoustic and text data. The late fusion of the best models outperforms the acoustic and text baselines by joint recognition balance metrics of 12.3% and 10.4%, respectively.

cs.SD↗

arXiv2025

Derailer-Rerailer: Adaptive Verification for Efficient and Reliable Language Model Reasoning

Large Language Models (LLMs) have shown impressive reasoning capabilities, yet existing prompting methods face a critical trade-off: simple approaches often struggle with complex tasks and reasoning stability, while more sophisticated methods require multiple inferences and substantial computational resources, limiting their practical deployment. To address this challenge, we propose Derailer-Rerailer, a novel framework that adaptively balances reasoning accuracy and computational efficiency. At its core, our framework employs a lightweight Derailer mechanism to assess reasoning stability and selectively triggers an advanced Rerailer verification process only when necessary, thereby optimizing computational resource usage. Extensive evaluation across both open and closed-source models on more than 20 categories of mathematical, symbolic, and commonsense reasoning tasks demonstrates our framework's effectiveness: Derailer-Rerailer achieves significant accuracy improvements (8-11\% across various reasoning tasks) while maintaining 2-3 times better efficiency than existing verification methods, with particularly strong performance in mathematical and symbolic reasoning, offering a practical solution for enhancing LLM reasoning reliability while significantly reducing computational overhead.

cs.CL↗

arXiv2025

A Unified Speech LLM for Diarization and Speech Recognition in Multilingual Conversations

Speech Large Language Models (Speech LLMs) have emerged as a crucial paradigm in recent years, extending the capabilities of traditional LLMs to speech tasks such as automatic speech recognition (ASR) and spoken dialogue modeling. However, their effectiveness in real-world multilingual conversations remains limited by the scarcity of data that captures natural conversational phenomena. To address this, the MLC-SLM Challenge provides a multilingual conversational dataset and evaluates models on two tasks: ASR with oracle segmentation (Task I) and joint diarization and recognition without oracle information (Task II). In this paper, we focus on Task II and propose a unified speech LLM that jointly performs diarization and ASR in an end-to-end manner. By reformulating the training data format and modifying the inference procedure, our model addresses the ambiguity inherent in pre-segmented audio and achieves a 54.87\% relative improvement in tcpWER/tcpCER over the baseline, ranking 8th overall, despite using a smaller LLM backbone. We also report results from Task I using a fine-tuned speech LLM.

cs.CL↗

arXiv2025

CmFNet: Cross-modal Fusion Network for Weakly-supervised Segmentation of Medical Images

Accurate automatic medical image segmentation relies on high-quality, dense annotations, which are costly and time-consuming. Weakly supervised learning provides a more efficient alternative by leveraging sparse and coarse annotations instead of dense, precise ones. However, segmentation performance degradation and overfitting caused by sparse annotations remain key challenges. To address these issues, we propose CmFNet, a novel 3D weakly supervised cross-modal medical image segmentation approach. CmFNet consists of three main components: a modality-specific feature learning network, a cross-modal feature learning network, and a hybrid-supervised learning strategy. Specifically, the modality-specific feature learning network and the cross-modal feature learning network effectively integrate complementary information from multi-modal images, enhancing shared features across modalities to improve segmentation performance. Additionally, the hybrid-supervised learning strategy guides segmentation through scribble supervision, intra-modal regularization, and inter-modal consistency, modeling spatial and contextual relationships while promoting feature alignment. Our approach effectively mitigates overfitting, delivering robust segmentation results. It excels in segmenting both challenging small tumor regions and common anatomical structures. Extensive experiments on a clinical cross-modal nasopharyngeal carcinoma (NPC) dataset (including CT and MR imaging) and the publicly available CT Whole Abdominal Organ dataset (WORD) show that our approach outperforms state-of-the-art weakly supervised methods. In addition, our approach also outperforms fully supervised methods when full annotation is used. Our approach can facilitate clinical therapy and benefit various specialists, including physicists, radiologists, pathologists, and oncologists.

cs.CV↗

arXiv2025

Semi-Supervised Multi-Modal Medical Image Segmentation for Complex Situations

Semi-supervised learning addresses the issue of limited annotations in medical images effectively, but its performance is often inadequate for complex backgrounds and challenging tasks. Multi-modal fusion methods can significantly improve the accuracy of medical image segmentation by providing complementary information. However, they face challenges in achieving significant improvements under semi-supervised conditions due to the challenge of effectively leveraging unlabeled data. There is a significant need to create an effective and reliable multi-modal learning strategy for leveraging unlabeled data in semi-supervised segmentation. To address these issues, we propose a novel semi-supervised multi-modal medical image segmentation approach, which leverages complementary multi-modal information to enhance performance with limited labeled data. Our approach employs a multi-stage multi-modal fusion and enhancement strategy to fully utilize complementary multi-modal information, while reducing feature discrepancies and enhancing feature sharing and alignment. Furthermore, we effectively introduce contrastive mutual learning to constrain prediction consistency across modalities, thereby facilitating the robustness of segmentation results in semi-supervised tasks. Experimental results on two multi-modal datasets demonstrate the superior performance and robustness of the proposed framework, establishing its valuable potential for solving medical image segmentation tasks in complex scenarios.

cs.CV↗

arXiv2025

Language-Aware Prompt Tuning for Parameter-Efficient Seamless Language Expansion in Multilingual ASR

Recent advancements in multilingual automatic speech recognition (ASR) have been driven by large-scale end-to-end models like Whisper. However, challenges such as language interference and expanding to unseen languages (language expansion) without degrading performance persist. This paper addresses these with three contributions: 1) Entire Soft Prompt Tuning (Entire SPT), which applies soft prompts to both the encoder and decoder, enhancing feature extraction and decoding; 2) Language-Aware Prompt Tuning (LAPT), which leverages cross-lingual similarities to encode shared and language-specific features using lightweight prompt matrices; 3) SPT-Whisper, a toolkit that integrates SPT into Whisper and enables efficient continual learning. Experiments across three languages from FLEURS demonstrate that Entire SPT and LAPT outperform Decoder SPT by 5.0% and 16.0% in language expansion tasks, respectively, providing an efficient solution for dynamic, multilingual ASR models with minimal computational overhead.

cs.CL↗

arXiv2025

Adapting Whisper for Parameter-efficient Code-Switching Speech Recognition via Soft Prompt Tuning

Large-scale multilingual ASR models like Whisper excel in high-resource settings but face challenges in low-resource scenarios, such as rare languages and code-switching (CS), due to computational costs and catastrophic forgetting. We explore Soft Prompt Tuning (SPT), a parameter-efficient method to enhance CS ASR while preserving prior knowledge. We evaluate two strategies: (1) full fine-tuning (FFT) of both soft prompts and the entire Whisper model, demonstrating improved cross-lingual capabilities compared to traditional methods, and (2) adhering to SPT's original design by freezing model parameters and only training soft prompts. Additionally, we introduce SPT4ASR, a combination of different SPT variants. Experiments on the SEAME and ASRU2019 datasets show that deep prompt tuning is the most effective SPT approach, and our SPT4ASR methods achieve further error reductions in CS ASR, maintaining parameter efficiency similar to LoRA, without degrading performance on existing languages.

cs.CL↗

arXiv2025

The Security Overview and Analysis of 3GPP 5G MAC CE

To more effectively control and allocate network resources, MAC CE has been introduced into the network protocol, which is a type of control signaling located in the MAC layer. Since MAC CE lacks encryption and integrity protection mechanisms provided by PDCP, the control signaling carried by MAC CE is vulnerable to interception or tampering by attackers during resource scheduling and allocation. Currently, the 3GPP has analyzed the security risks of Layer 1/Layer 2 Triggered Mobility (LTM), where handover signaling sent to the UE via MAC CE by the network can lead to privacy leaks and network attacks. However, in addition to LTM, there may be other potential security vulnerabilities in other protocol procedures. Therefore, this paper explores the security threats to MAC CE and the corresponding protection mechanisms. The research is expected to support the 3GPP's study of MAC CE and be integrated with the security research of lower-layer protocols, thereby enhancing the security and reliability of the entire communication system.

cs.CR↗

arXiv2025

Stroke-based Cyclic Amplifier: Image Super-Resolution at Arbitrary Ultra-Large Scales

Prior Arbitrary-Scale Image Super-Resolution (ASISR) methods often experience a significant performance decline when the upsampling factor exceeds the range covered by the training data, introducing substantial blurring. To address this issue, we propose a unified model, Stroke-based Cyclic Amplifier (SbCA), for ultra-large upsampling tasks. The key of SbCA is the stroke vector amplifier, which decomposes the image into a series of strokes represented as vector graphics for magnification. Then, the detail completion module also restores missing details, ensuring high-fidelity image reconstruction. Our cyclic strategy achieves ultra-large upsampling by iteratively refining details with this unified SbCA model, trained only once for all, while keeping sub-scales within the training range. Our approach effectively addresses the distribution drift issue and eliminates artifacts, noise and blurring, producing high-quality, high-resolution super-resolved images. Experimental validations on both synthetic and real-world datasets demonstrate that our approach significantly outperforms existing methods in ultra-large upsampling tasks (e.g. $\times100$), delivering visual quality far superior to state-of-the-art techniques.

cs.CV↗

arXiv2025

BalancEdit: Dynamically Balancing the Generality-Locality Trade-off in Multi-modal Model Editing

Large multi-modal models inevitably decay over time as facts update and previously learned information becomes outdated. Traditional approaches such as fine-tuning are often impractical for updating these models due to their size and complexity. Instead, direct knowledge editing within the models presents a more viable solution. Current model editing techniques, however, typically overlook the unique influence ranges of different facts, leading to compromised model performance in terms of both generality and locality. To address this issue, we introduce the concept of the generality-locality trade-off in multi-modal model editing. We develop a new model editing dataset named OKEDIT, specifically designed to effectively evaluate this trade-off. Building on this foundation, we propose \textbf{BalancEdit}, a novel method for balanced model editing that dynamically achieves an optimal balance between generality and locality. BalancEdit utilizes a unique mechanism that generates both positive and negative samples for each fact to accurately determine its influence scope and incorporates these insights into the model's latent space using a discrete, localized codebook of edits, without modifying the underlying model weights. To our knowledge, this is the first approach explicitly addressing the generality-locality trade-off in multi-modal model editing. Our comprehensive results confirm the effectiveness of BalancEdit, demonstrating minimal trade-offs while maintaining robust editing capabilities. Our code and dataset are available at https://github.com/donglgcn/BalancEdit/tree/MMOKVQA.

cs.AI↗

arXiv2025

Query Nearby: Offset-Adjusted Mask2Former enhances small-organ segmentation

Medical segmentation plays an important role in clinical applications like radiation therapy and surgical guidance, but acquiring clinically acceptable results is difficult. In recent years, progress has been witnessed with the success of utilizing transformer-like models, such as combining the attention mechanism with CNN. In particular, transformer-based segmentation models can extract global information more effectively, compensating for the drawbacks of CNN modules that focus on local features. However, utilizing transformer architecture is not easy, because training transformer-based models can be resource-demanding. Moreover, due to the distinct characteristics in the medical field, especially when encountering mid-sized and small organs with compact regions, their results often seem unsatisfactory. For example, using ViT to segment medical images directly only gives a DSC of less than 50\%, which is far lower than the clinically acceptable score of 80\%. In this paper, we used Mask2Former with deformable attention to reduce computation and proposed offset adjustment strategies to encourage sampling points within the same organs during attention weights computation, thereby integrating compact foreground information better. Additionally, we utilized the 4th feature map in Mask2Former to provide a coarse location of organs, and employed an FCN-based auxiliary head to help train Mask2Former more quickly using Dice loss. We show that our model achieves SOTA (State-of-the-Art) performance on the HaNSeg and SegRap2023 datasets, especially on mid-sized and small organs.Our code is available at link https://github.com/earis/Offsetadjustment\_Background-location\_Decoder\_Mask2former.

cs.CV↗

arXiv2025

NAT: Neural Acoustic Transfer for Interactive Scenes in Real Time

Previous acoustic transfer methods rely on extensive precomputation and storage of data to enable real-time interaction and auditory feedback. However, these methods struggle with complex scenes, especially when dynamic changes in object position, material, and size significantly alter sound effects. These continuous variations lead to fluctuating acoustic transfer distributions, making it challenging to represent with basic data structures and render efficiently in real time. To address this challenge, we present Neural Acoustic Transfer, a novel approach that utilizes an implicit neural representation to encode precomputed acoustic transfer and its variations, allowing for real-time prediction of sound fields under varying conditions. To efficiently generate the training data required for the neural acoustic field, we developed a fast Monte-Carlo-based boundary element method (BEM) approximation for general scenarios with smooth Neumann conditions. Additionally, we implemented a GPU-accelerated version of standard BEM for scenarios requiring higher precision. These methods provide the necessary training data, enabling our neural network to accurately model the sound radiation space. We demonstrate our method's numerical accuracy and runtime efficiency (within several milliseconds for 30s audio) through comprehensive validation and comparisons in diverse acoustic transfer scenarios. Our approach allows for efficient and accurate modeling of sound behavior in dynamically changing environments, which can benefit a wide range of interactive applications such as virtual reality, augmented reality, and advanced audio production.

cs.SD↗

arXiv2025

RoE-FND: A Case-Based Reasoning Approach with Dual Verification for Fake News Detection via LLMs

The proliferation of deceptive content online necessitates robust Fake News Detection (FND) systems. While evidence-based approaches leverage external knowledge to verify claims, existing methods face critical limitations: noisy evidence selection, generalization bottlenecks, and unclear decision-making processes. Recent efforts to harness Large Language Models (LLMs) for FND introduce new challenges, including hallucinated rationales and conclusion bias. To address these issues, we propose \textbf{RoE-FND} (\textbf{\underline{R}}eason \textbf{\underline{o}}n \textbf{\underline{E}}xperiences FND), a framework that reframes evidence-based FND as a logical deduction task by synergizing LLMs with experiential learning. RoE-FND encompasses two stages: (1) \textit{self-reflective knowledge building}, where a knowledge base is curated by analyzing past reasoning errors, namely the exploration stage, and (2) \textit{dynamic criterion retrieval}, which synthesizes task-specific reasoning guidelines from historical cases as experiences during deployment. It further cross-checks rationales against internal experience through a devised dual-channel procedure. Key contributions include: a case-based reasoning framework for FND that addresses multiple existing challenges, a training-free approach enabling adaptation to evolving situations, and empirical validation of the framework's superior generalization and effectiveness over state-of-the-art methods across three datasets.

cs.CL↗

arXiv2025

Benign Samples Matter! Fine-tuning On Outlier Benign Samples Severely Breaks Safety

Recent studies have uncovered a troubling vulnerability in the fine-tuning stage of large language models (LLMs): even fine-tuning on entirely benign datasets can lead to a significant increase in the harmfulness of LLM outputs. Building on this finding, our red teaming study takes this threat one step further by developing a more effective attack. Specifically, we analyze and identify samples within benign datasets that contribute most to safety degradation, then fine-tune LLMs exclusively on these samples. We approach this problem from an outlier detection perspective and propose Self-Inf-N, to detect and extract outliers for fine-tuning. Our findings reveal that fine-tuning LLMs on 100 outlier samples selected by Self-Inf-N in the benign datasets severely compromises LLM safety alignment. Extensive experiments across seven mainstream LLMs demonstrate that our attack exhibits high transferability across different architectures and remains effective in practical scenarios. Alarmingly, our results indicate that most existing mitigation strategies fail to defend against this attack, underscoring the urgent need for more robust alignment safeguards. Codes are available at https://github.com/GuanZihan/Benign-Samples-Matter.

cs.LG↗

arXiv2025

CoTSRF: Utilize Chain of Thought as Stealthy and Robust Fingerprint of Large Language Models

Despite providing superior performance, open-source large language models (LLMs) are vulnerable to abusive usage. To address this issue, recent works propose LLM fingerprinting methods to identify the specific source LLMs behind suspect applications. However, these methods fail to provide stealthy and robust fingerprint verification. In this paper, we propose a novel LLM fingerprinting scheme, namely CoTSRF, which utilizes the Chain of Thought (CoT) as the fingerprint of an LLM. CoTSRF first collects the responses from the source LLM by querying it with crafted CoT queries. Then, it applies contrastive learning to train a CoT extractor that extracts the CoT feature (i.e., fingerprint) from the responses. Finally, CoTSRF conducts fingerprint verification by comparing the Kullback-Leibler divergence between the CoT features of the source and suspect LLMs against an empirical threshold. Various experiments have been conducted to demonstrate the advantage of our proposed CoTSRF for fingerprinting LLMs, particularly in stealthy and robust fingerprint verification.

cs.CR↗

arXiv2025

EdgeOL: Efficient in-situ Online Learning on Edge Devices

Emerging applications, such as robot-assisted eldercare and object recognition, generally employ deep learning neural networks (DNNs) and naturally require: i) handling streaming-in inference requests and ii) adapting to possible deployment scenario changes. Online model fine-tuning is widely adopted to satisfy these needs. However, an inappropriate fine-tuning scheme could involve significant energy consumption, making it challenging to deploy on edge devices. In this paper, we propose EdgeOL, an edge online learning framework that optimizes inference accuracy, fine-tuning execution time, and energy efficiency through both inter-tuning and intra-tuning optimizations. Experimental results show that, on average, EdgeOL reduces overall fine-tuning execution time by 64%, energy consumption by 52%, and improves average inference accuracy by 1.75% over the immediate online learning strategy

cs.LG↗

arXiv2025

Adaptive Mixture of Low-Rank Experts for Robust Audio Spoofing Detection

In audio spoofing detection, most studies rely on clean datasets, making models susceptible to real-world post-processing attacks, such as channel compression and noise. To overcome this challenge, we propose the Adaptive MixtUre Low-rank ExperTs (AMULET) framework, which enhances resilience by leveraging attack-specific knowledge and dynamically adapting to varied attack conditions. Specifically, AMULET employs Attack-Specific Experts (ASEs) fine-tuned with Low-Rank Adaptation (LoRA), allowing each expert to focus on distinct post-processing patterns using just 1.13\% of the parameters required for full fine-tuning. Furthermore, we introduce Adaptive Expert Fusion (AEF), which adaptively selects and integrates expert knowledge to enhance the robustness of spoofing detection. Experimental results demonstrate that AMULET significantly enhances robustness by improving noise resilience and exhibiting greater adaptability to unseen post-processing methods compared to models trained with full fine-tuning. Additionally, our framework outperforms both single expert and other expert aggregation strategies under various mixed attacks, demonstrating its superior robustness and adaptability in managing complex real-world scenarios.

eess.AS↗