Wenxuan Wang

Wenxiang Jiao, Jen-tse Huang, Wenxuan Wang et al. · pku, tencent-ai

Large language models (LLMs) like ChatGPT have exhibited remarkable abilities on a wide range of natural language processing~(NLP) tasks, including various machine translation abilities accomplished during chat. However, these models are only accessible through restricted APIs, which creates barriers to new research and advancements in the field. Therefore, we propose ParroT, a framework to enhance and regulate the translation abilities during chat based on open-source LLMs (e.g., LLaMA), human-written translation and feedback data. Specifically, ParroT reformulates translation data into the instruction-following style, and introduces a "$\mathbf{Hint}$" field for incorporating extra requirements to regulate the translation process. Accordingly, we propose three instruction types for finetuning ParroT models, including translation instruction, contrastive instruction, and error-guided instruction. Experiments on Flores subsets and WMT22 test sets suggest that translation instruction improves the translation performance of vanilla LLMs significantly while error-guided instruction can lead to further improvement, which demonstrates the importance of learning from low-quality translations annotated by humans. We also demonstrate the potential of automatic evaluation tools in providing quality information of translations, when constructing error-guided instructions for directions that lack human annotation data. Please refer to our Github project for more implementation details: https://github.com/wxjiao/ParroT

Ada Chen, Yongjiang Wu, Junyuan Zhang et al. · pku, tencent-ai

Recently, AI-driven interactions with computing devices have advanced from basic prototype tools to sophisticated, LLM-based systems that emulate human-like operations in graphical user interfaces. We are now witnessing the emergence of \emph{Computer-Using Agents} (CUAs), capable of autonomously performing tasks such as navigating desktop applications, web pages, and mobile apps. However, as these agents grow in capability, they also introduce novel safety and security risks. Vulnerabilities in LLM-driven reasoning, with the added complexity of integrating multiple software components and multimodal inputs, further complicate the security landscape. In this paper, we present a systematization of knowledge on the safety and security threats of CUAs. We conduct a comprehensive literature review and distill our findings along four research objectives: \textit{\textbf{(i)}} define the CUA that suits safety analysis; \textit{\textbf{(ii)} } categorize current safety threats among CUAs; \textit{\textbf{(iii)}} propose a comprehensive taxonomy of existing defensive strategies; \textit{\textbf{(iv)}} summarize prevailing benchmarks, datasets, and evaluation metrics used to assess the safety and performance of CUAs. Building on these insights, our work provides future researchers with a structured foundation for exploring unexplored vulnerabilities and offers practitioners actionable guidance in designing and deploying secure Computer-Using Agents.

Wenxuan Wang, Wenxiang Jiao, Jingyuan Huang et al. · pku, tencent-ai

This paper identifies a cultural dominance issue within large language models (LLMs) due to the predominant use of English data in model training (e.g., ChatGPT). LLMs often provide inappropriate English-culture-related answers that are not relevant to the expected culture when users ask in non-English languages. To systematically evaluate the cultural dominance issue, we build a benchmark of concrete (e.g., holidays and songs) and abstract (e.g., values and opinions) cultural objects. Empirical results show that the representative GPT models suffer from the culture dominance problem, where GPT-4 is the most affected while text-davinci-003 suffers the least from this problem. Our study emphasizes the need to critically examine cultural dominance and ethical consideration in their development and deployment. We show that two straightforward methods in model development (i.e., pretraining on more diverse data) and deployment (e.g., culture-aware prompting) can significantly mitigate the cultural dominance issue in LLMs.

Wenxuan Wang, Juluan Shi, Zixuan Ling et al. · pku, tencent-ai

Equipped with the capability to call functions, modern large language models (LLMs) can leverage external tools for addressing a range of tasks unattainable through language skills alone. However, the effective execution of these tools relies heavily not just on the advanced capabilities of LLMs but also on precise user instructions, which often cannot be ensured in the real world. To evaluate the performance of LLMs tool-use under imperfect instructions, we meticulously examine the real-world instructions queried from users, analyze the error patterns, and build a challenging tool-use benchmark called Noisy ToolBench (NoisyToolBench). We find that due to the next-token prediction training objective, LLMs tend to arbitrarily generate the missed argument, which may lead to hallucinations and risks. To address this issue, we propose a novel framework, Ask-when-Needed (AwN), which prompts LLMs to ask questions to users whenever they encounter obstacles due to unclear instructions. Moreover, to reduce the manual labor involved in user-LLM interaction and assess LLMs performance in tool utilization from both accuracy and efficiency perspectives, we design an automated evaluation tool named ToolEvaluator. Our experiments demonstrate that the AwN significantly outperforms existing frameworks for tool learning in the NoisyToolBench. We will release all related code and datasets to support future research.

Tian Liang, Zhiwei He, Jen-tse Huang et al. · pku, tencent-ai

The automatic evaluation of LLM-based agent intelligence is critical in developing advanced LLM-based agents. Although considerable effort has been devoted to developing human-annotated evaluation datasets, such as AlpacaEval, existing techniques are costly, time-consuming, and lack adaptability. In this paper, inspired by the popular language game ``Who is Spy'', we propose to use the word guessing game to assess the intelligence performance of LLMs. Given a word, the LLM is asked to describe the word and determine its identity (spy or not) based on its and other players' descriptions. Ideally, an advanced agent should possess the ability to accurately describe a given word using an aggressive description while concurrently maximizing confusion in the conservative description, enhancing its participation in the game. To this end, we first develop DEEP to evaluate LLMs' expression and disguising abilities. DEEP requires LLM to describe a word in aggressive and conservative modes. We then introduce SpyGame, an interactive multi-agent framework designed to assess LLMs' intelligence through participation in a competitive language-based board game. Incorporating multi-agent interaction, SpyGame requires the target LLM to possess linguistic skills and strategic thinking, providing a more comprehensive evaluation of LLMs' human-like cognitive abilities and adaptability in complex communication situations. The proposed evaluation framework is very easy to implement. We collected words from multiple sources, domains, and languages and used the proposed evaluation framework to conduct experiments. Extensive experiments demonstrate that the proposed DEEP and SpyGame effectively evaluate the capabilities of various LLMs, capturing their ability to adapt to novel situations and engage in strategic communication.

Wenxuan Wang, Jingyuan Huang, Jen-tse Huang et al. · pku, tencent-ai

The exponential growth of social media platforms has brought about a revolution in communication and content dissemination in human society. Nevertheless, these platforms are being increasingly misused to spread toxic content, including hate speech, malicious advertising, and pornography, leading to severe negative consequences such as harm to teenagers' mental health. Despite tremendous efforts in developing and deploying textual and image content moderation methods, malicious users can evade moderation by embedding texts into images, such as screenshots of the text, usually with some interference. We find that modern content moderation software's performance against such malicious inputs remains underexplored. In this work, we propose OASIS, a metamorphic testing framework for content moderation software. OASIS employs 21 transform rules summarized from our pilot study on 5,000 real-world toxic contents collected from 4 popular social media applications, including Twitter, Instagram, Sina Weibo, and Baidu Tieba. Given toxic textual contents, OASIS can generate image test cases, which preserve the toxicity yet are likely to bypass moderation. In the evaluation, we employ OASIS to test five commercial textual content moderation software from famous companies (i.e., Google Cloud, Microsoft Azure, Baidu Cloud, Alibaba Cloud and Tencent Cloud), as well as a state-of-the-art moderation research model. The results show that OASIS achieves up to 100% error finding rates. Moreover, through retraining the models with the test cases generated by OASIS, the robustness of the moderation model can be improved without performance degradation.

Xujie Song, Tong Liu, Shengbo Eben Li et al.

As a dedicated quantum device, Ising machines could solve large-scale binary optimization problems in milliseconds. There is emerging interest in utilizing Ising machines to train feedforward neural networks due to the prosperity of generative artificial intelligence. However, existing methods can only train single-layer feedforward networks because of the complex nonlinear network topology. This paper proposes an Ising learning algorithm to train quantized neural network (QNN), by incorporating two essential techinques, namely binary representation of topological network and order reduction of loss function. As far as we know, this is the first algorithm to train multi-layer feedforward networks on Ising machines, providing an alternative to gradient-based backpropagation. Firstly, training QNN is formulated as a quadratic constrained binary optimization (QCBO) problem by representing neuron connection and activation function as equality constraints. All quantized variables are encoded by binary bits based on binary encoding protocol. Secondly, QCBO is converted to a quadratic unconstrained binary optimization (QUBO) problem, that can be efficiently solved on Ising machines. The conversion leverages both penalty function and Rosenberg order reduction, who together eliminate equality constraints and reduce high-order loss function into a quadratic one. With some assumptions, theoretical analysis shows the space complexity of our algorithm is $\mathcal{O}(H^2L + HLN\log H)$, quantifying the required number of Ising spins. Finally, the algorithm effectiveness is validated with a simulated Ising machine on MNIST dataset. After annealing 700 ms, the classification accuracy achieves 98.3%. Among 100 runs, the success probability of finding the optimal solution is 72%. Along with the increasing number of spins on Ising machine, our algorithm has the potential to train deeper neural networks.

Jen-tse Huang, Eric John Li, Man Ho Lam et al. · pku, tencent-ai

Decision-making is a complex process requiring diverse abilities, making it an excellent framework for evaluating Large Language Models (LLMs). Researchers have examined LLMs' decision-making through the lens of Game Theory. However, existing evaluation mainly focus on two-player scenarios where an LLM competes against another. Additionally, previous benchmarks suffer from test set leakage due to their static design. We introduce GAMA($γ$)-Bench, a new framework for evaluating LLMs' Gaming Ability in Multi-Agent environments. It includes eight classical game theory scenarios and a dynamic scoring scheme specially designed to quantitatively assess LLMs' performance. $γ$-Bench allows flexible game settings and adapts the scoring system to different game parameters, enabling comprehensive evaluation of robustness, generalizability, and strategies for improvement. Our results indicate that GPT-3.5 demonstrates strong robustness but limited generalizability, which can be enhanced using methods like Chain-of-Thought. We also evaluate 13 LLMs from 6 model families, including GPT-3.5, GPT-4, Gemini, LLaMA-3.1, Mixtral, and Qwen-2. Gemini-1.5-Pro outperforms others, scoring of $69.8$ out of $100$, followed by LLaMA-3.1-70B ($65.9$) and Mixtral-8x22B ($62.4$). Our code and experimental results are publicly available at https://github.com/CUHK-ARISE/GAMABench.

Jialun Cao, Yuk-Kit Chan, Zixuan Ling et al.

Various benchmarks have been proposed to assess the performance of large language models (LLMs) in different coding scenarios. We refer to them as code-related benchmarks. However, there are no systematic guidelines by which such a benchmark should be developed to ensure its quality, reliability, and reproducibility. We propose How2Bench, which is comprised of a 55-criteria checklist as a set of guidelines to govern the development of code-related benchmarks comprehensively. Using HOW2BENCH, we profiled 274 benchmarks released within the past decade and found concerning issues. Nearly 70% of the benchmarks did not take measures for data quality assurance; over 10% did not even open source or only partially open source. Many highly cited benchmarks have loopholes, including duplicated samples, incorrect reference codes/tests/prompts, and unremoved sensitive/confidential information. Finally, we conducted a human study involving 49 participants, which revealed significant gaps in awareness of the importance of data quality, reproducibility, and transparency.

Jingyuan Huang, Jen-tse Huang, Ziyi Liu et al. · pku, tencent-ai

Visual-Language Models (VLMs) have shown remarkable performance across various tasks, particularly in recognizing geographic information from images. However, VLMs still show regional biases in this task. To systematically evaluate these issues, we introduce a benchmark consisting of 1,200 images paired with detailed geographic metadata. Evaluating four VLMs, we find that while these models demonstrate the ability to recognize geographic information from images, achieving up to 53.8% accuracy in city prediction, they exhibit significant biases. Specifically, performance is substantially higher for economically developed and densely populated regions compared to less developed (-12.5%) and sparsely populated (-17.0%) areas. Moreover, regional biases of frequently over-predicting certain locations remain. For instance, they consistently predict Sydney for images taken in Australia, shown by the low entropy scores for these countries. The strong performance of VLMs also raises privacy concerns, particularly for users who share images online without the intent of being identified. Our code and dataset are publicly available at https://github.com/uscnlp-lime/FairLocator.

Wenxuan Wang, Yisi Zhang, Xingjian He et al.

Visual grounding (VG) aims at locating the foreground entities that match the given natural language expressions. Previous datasets and methods for classic VG task mainly rely on the prior assumption that the given expression must literally refer to the target object, which greatly impedes the practical deployment of agents in real-world scenarios. Since users usually prefer to provide intention-based expression for the desired object instead of covering all the details, it is necessary for the agents to interpret the intention-driven instructions. Thus, in this work, we take a step further to the intention-driven visual-language (V-L) understanding. To promote classic VG towards human intention interpretation, we propose a new intention-driven visual grounding (IVG) task and build a large-scale IVG dataset termed IntentionVG with free-form intention expressions. Considering that practical agents need to move and find specific targets among various scenarios to realize the grounding task, our IVG task and IntentionVG dataset have taken the crucial properties of both multi-scenario perception and egocentric view into consideration. Besides, various types of models are set up as the baselines to realize our IVG task. Extensive experiments on our IntentionVG dataset and baselines demonstrate the necessity and efficacy of our method for the V-L field. To foster future research in this direction, our newly built dataset and baselines will be publicly available at https://github.com/Rubics-Xuan/IVG.

Kunyu Wang, Juluan Shi, Wenxuan Wang

Deep neural networks are susceptible to adversarial attacks, which pose a significant threat to their security and reliability in real-world applications. The most notable adversarial attacks are transfer-based attacks, where an adversary crafts an adversarial example to fool one model, which can also fool other models. While previous research has made progress in improving the transferability of untargeted adversarial examples, the generation of targeted adversarial examples that can transfer between models remains a challenging task. In this work, we present a novel approach to generate transferable targeted adversarial examples by exploiting the vulnerability of deep neural networks to perturbations on high-frequency components of images. We observe that replacing the high-frequency component of an image with that of another image can mislead deep models, motivating us to craft perturbations containing high-frequency information to achieve targeted attacks. To this end, we propose a method called Low-Frequency Adversarial Attack (\name), which trains a conditional generator to generate targeted adversarial perturbations that are then added to the low-frequency component of the image. Extensive experiments on ImageNet demonstrate that our proposed approach significantly outperforms state-of-the-art methods, improving targeted attack success rates by a margin from 3.2\% to 15.5\%.

Jen-tse Huang, Yuhang Yan, Linqi Liu et al. · pku, tencent-ai

Recent failures such as Google Gemini generating people of color in Nazi-era uniforms illustrate how AI outputs can be factually plausible yet socially harmful. AI models are increasingly evaluated for "fairness," yet existing benchmarks often conflate two fundamentally different dimensions: factual correctness and normative fairness. A model may generate responses that are factually accurate but socially unfair, or conversely, appear fair while distorting factual reality. We argue that identifying the boundary between fact and fair is essential for meaningful fairness evaluation. We introduce Fact-or-Fair, a benchmark with (i) objective queries aligned with descriptive, fact-based judgments, and (ii) subjective queries aligned with normative, fairness-based judgments. Our queries are constructed from 19 statistics and are grounded in cognitive psychology, drawing on representativeness bias, attribution bias, and ingroup-outgroup bias to explain why models often misalign fact and fairness. Experiments across ten frontier models reveal different levels of fact-fair trade-offs. By reframing fairness evaluation, we provide both a new theoretical lens and a practical benchmark to advance the responsible model assessments. Our test suite is publicly available at https://github.com/uclanlp/Fact-or-Fair.

Wenxuan Wang, Haonan Bai, Jen-tse Huang et al. · pku, tencent-ai

Image generation models can generate or edit images from a given text. Recent advancements in image generation technology, exemplified by DALL-E and Midjourney, have been groundbreaking. These advanced models, despite their impressive capabilities, are often trained on massive Internet datasets, making them susceptible to generating content that perpetuates social stereotypes and biases, which can lead to severe consequences. Prior research on assessing bias within image generation models suffers from several shortcomings, including limited accuracy, reliance on extensive human labor, and lack of comprehensive analysis. In this paper, we propose BiasPainter, a novel evaluation framework that can accurately, automatically and comprehensively trigger social bias in image generation models. BiasPainter uses a diverse range of seed images of individuals and prompts the image generation models to edit these images using gender, race, and age-neutral queries. These queries span 62 professions, 39 activities, 57 types of objects, and 70 personality traits. The framework then compares the edited images to the original seed images, focusing on the significant changes related to gender, race, and age. BiasPainter adopts a key insight that these characteristics should not be modified when subjected to neutral prompts. Built upon this design, BiasPainter can trigger the social bias and evaluate the fairness of image generation models. We use BiasPainter to evaluate six widely-used image generation models, such as stable diffusion and Midjourney. Experimental results show that BiasPainter can successfully trigger social bias in image generation models. According to our human evaluation, BiasPainter can achieve 90.8% accuracy on automatic bias detection, which is significantly higher than the results reported in previous work.

Yuxuan Wan, Yi Dong, Jingyu Xiao et al.

Multi-page websites dominate modern web development. However, existing design-to-code methods rely on simplified assumptions, limiting to single-page, self-contained webpages without external resource connection. To address this gap, we introduce the Multi-Page Resource-Aware Webpage (MRWeb) generation task, which transforms UI designs into multi-page, functional web UIs with internal/external navigation, image loading, and backend routing. We propose a novel resource list data structure to track resources, links, and design components. Our study applies existing methods to the MRWeb problem using a newly curated dataset of 500 websites (300 synthetic, 200 real-world). Specifically, we identify the best metric to evaluate the similarity of the web UI, assess the impact of the resource list on MRWeb generation, analyze MLLM limitations, and evaluate the effectiveness of the MRWeb tool in real-world workflows. The results show that resource lists boost navigation functionality from 0% to 66%-80% while facilitating visual similarity. Our proposed metrics and evaluation framework provide new insights into MLLM performance on MRWeb tasks. We release the MRWeb tool, dataset, and evaluation framework to promote further research.

Fengfan Zhou, Bangjie Yin, Hefei Ling et al.

Face Recognition (FR) models are vulnerable to adversarial examples that subtly manipulate benign face images, underscoring the urgent need to improve the transferability of adversarial attacks in order to expose the blind spots of these systems. Existing adversarial attack methods often overlook the potential benefits of augmenting the surrogate model with diverse initializations, which limits the transferability of the generated adversarial examples. To address this gap, we propose a novel method called Diverse Parameters Augmentation (DPA) attack method, which enhances surrogate models by incorporating diverse parameter initializations, resulting in a broader and more diverse set of surrogate models. Specifically, DPA consists of two key stages: Diverse Parameters Optimization (DPO) and Hard Model Aggregation (HMA). In the DPO stage, we initialize the parameters of the surrogate model using both pre-trained and random parameters. Subsequently, we save the models in the intermediate training process to obtain a diverse set of surrogate models. During the HMA stage, we enhance the feature maps of the diversified surrogate models by incorporating beneficial perturbations, thereby further improving the transferability. Experimental results demonstrate that our proposed attack method can effectively enhance the transferability of the crafted adversarial face examples.

Wenxuan Wang, Juluan Shi, Zhaopeng Tu et al. · pku, tencent-ai

Large Language Models (LLMs) like ChatGPT are foundational in various applications due to their extensive knowledge from pre-training and fine-tuning. Despite this, they are prone to generating factual and commonsense errors, raising concerns in critical areas like healthcare, journalism, and education to mislead users. Current methods for evaluating LLMs' veracity are limited by test data leakage or the need for extensive human labor, hindering efficient and accurate error detection. To tackle this problem, we introduce a novel, automatic testing framework, FactChecker, aimed at uncovering factual inaccuracies in LLMs. This framework involves three main steps: First, it constructs a factual knowledge graph by retrieving fact triplets from a large-scale knowledge database. Then, leveraging the knowledge graph, FactChecker employs a rule-based approach to generates three types of questions (Yes-No, Multiple-Choice, and WH questions) that involve single-hop and multi-hop relations, along with correct answers. Lastly, it assesses the LLMs' responses for accuracy using tailored matching strategies for each question type. Our extensive tests on six prominent LLMs, including text-davinci-002, text-davinci-003, ChatGPT~(gpt-3.5-turbo, gpt-4), Vicuna, and LLaMA-2, reveal that FactChecker can trigger factual errors in up to 45\% of questions in these models. Moreover, we demonstrate that FactChecker's test cases can improve LLMs' factual accuracy through in-context learning and fine-tuning (e.g., llama-2-13b-chat's accuracy increase from 35.3\% to 68.5\%). We are making all code, data, and results available for future research endeavors.

Wenxuan Wang, Wenxiang Jiao, Jen-tse Huang et al. · pku, tencent-ai

In this study, we revisit the commonly-cited off-target issue in multilingual neural machine translation (MNMT). By carefully designing experiments on different MNMT scenarios and models, we attribute the off-target issue to the overfitting of the shortcuts of (non-centric, centric) language mappings. Specifically, the learned shortcuts biases MNMT to mistakenly translate non-centric languages into the centric language instead of the expected non-centric language for zero-shot translation. Analyses on learning dynamics show that the shortcut learning generally occurs in the later stage of model training, and multilingual pretraining accelerates and aggravates the shortcut learning. Based on these observations, we propose a simple and effective training strategy to eliminate the shortcuts in MNMT models by leveraging the forgetting nature of model training. The only difference from the standard training is that we remove the training instances that may induce the shortcut learning in the later stage of model training. Without introducing any additional data and computational costs, our approach can consistently and significantly improve the zero-shot translation performance by alleviating the shortcut learning for different MNMT models and benchmarks.

Youliang Yuan, Qiuyang Mang, Jingbang Chen et al. · pku, tencent-ai

Large language models for mathematical reasoning are typically trained with outcome-based rewards, which credit only the final answer. In our experiments, we observe that this paradigm is highly susceptible to reward hacking, leading to a substantial overestimation of a model's reasoning ability. This is evidenced by a high incidence of false positives - solutions that reach the correct final answer through an unsound reasoning process. Through a systematic analysis with human verification, we establish a taxonomy of these failure modes, identifying patterns like Miracle Steps - abrupt jumps to a correct output without a valid preceding derivation. Probing experiments suggest a strong association between these Miracle Steps and memorization, where the model appears to recall the answer directly rather than deriving it. To mitigate this systemic issue, we introduce the Rubric Reward Model (RRM), a process-oriented reward function that evaluates the entire reasoning trajectory against problem-specific rubrics. The generative RRM provides fine-grained, calibrated rewards (0-1) that explicitly penalize logical flaws and encourage rigorous deduction. When integrated into a reinforcement learning pipeline, RRM-based training consistently outperforms outcome-only supervision across four math benchmarks. Notably, it boosts Verified Pass@1024 on AIME2024 from 26.7% to 62.6% and reduces the incidence of Miracle Steps by 71%. Our work demonstrates that rewarding the solution process is crucial for building models that are not only more accurate but also more reliable.

Zhengliang Shi, Ruotian Ma, Jen-tse Huang et al. · pku, tencent-ai

Large language models (LLMs) are increasingly entrusted with high-stakes decisions that affect human welfare. However, the principles and values that guide these models when distributing scarce societal resources remain largely unexamined. To address this, we introduce the Social Welfare Function (SWF) Benchmark, a dynamic simulation environment where an LLM acts as a sovereign allocator, distributing tasks to a heterogeneous community of recipients. The benchmark is designed to create a persistent trade-off between maximizing collective efficiency (measured by Return on Investment) and ensuring distributive fairness (measured by the Gini coefficient). We evaluate 20 state-of-the-art LLMs and present the first leaderboard for social welfare allocation. Our findings reveal three key insights: (i) A model's general conversational ability, as measured by popular leaderboards, is a poor predictor of its allocation skill. (ii) Most LLMs exhibit a strong default utilitarian orientation, prioritizing group productivity at the expense of severe inequality. (iii) Allocation strategies are highly vulnerable, easily perturbed by output-length constraints and social-influence framing. These results highlight the risks of deploying current LLMs as societal decision-makers and underscore the need for specialized benchmarks and targeted alignment for AI governance.

Peng Du, Jinjing Shi, Wenxuan Wang et al.

Attention mechanisms underpin modern deep learning, while the quadratic time and space complexity limit scalability for long sequences. To address this, Quantum Annealing Multi-Head Attention (QAMA) is proposed, a novel drop-in operator that reformulates attention as an energy-based Hamiltonian optimization problem. In this framework, token interactions are encoded into binary quadratic terms, and quantum annealing is employed to search for low-energy configurations that correspond to effective attention patterns. Unlike classical sparse or approximate attention methods that rely on hand-crafted heuristics, QAMA allows sparsity structures to emerge naturally from the optimization process. Theoretically, computational complexity is analysed through single-spin flip dynamics, providing time to solution runtime bounds that depend on the spectral properties of the annealing Hamiltonian. Empirically, evaluation on both natural language and vision benchmarks shows that, across tasks, accuracy deviates by at most 2.7 points from standard multi-head attention, while requiring only linear qubits in sequence length. Visualizations further reveal that the Hamiltonian penalty terms induce meaningful and interpretable sparsity across heads. Finally, deployment on a coherent Ising machine validates the feasibility of running QAMA on real quantum hardware, showing tangible inference-time reductions compared with classical implementations. These results highlight QAMA as a pioneering and scalable step toward integrating quantum optimization devices into deep neural architectures, providing a seamlessly integrable and hardware-compatible alternative to conventional attention mechanisms. This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.

Jen-tse Huang, Wenxiang Jiao, Man Ho Lam et al.

Recent research has focused on examining Large Language Models' (LLMs) characteristics from a psychological standpoint, acknowledging the necessity of understanding their behavioral characteristics. The administration of personality tests to LLMs has emerged as a noteworthy area in this context. However, the suitability of employing psychological scales, initially devised for humans, on LLMs is a matter of ongoing debate. Our study aims to determine the reliability of applying personality assessments to LLMs, explicitly investigating whether LLMs demonstrate consistent personality traits. Analysis of 2,500 settings per model, including GPT-3.5, GPT-4, Gemini-Pro, and LLaMA-3.1, reveals that various LLMs show consistency in responses to the Big Five Inventory, indicating a satisfactory level of reliability. Furthermore, our research explores the potential of GPT-3.5 to emulate diverse personalities and represent various groups-a capability increasingly sought after in social sciences for substituting human participants with LLMs to reduce costs. Our findings reveal that LLMs have the potential to represent different personalities with specific prompt instructions.

Wenxuan Wang, Jingyuan Huang, Chang Chen et al.

The exponential growth of social media platforms, such as Facebook and TikTok, has revolutionized communication and content publication in human society. Users on these platforms can publish multimedia content that delivers information via the combination of text, audio, images, and video. Meanwhile, the multimedia content release facility has been increasingly exploited to propagate toxic content, such as hate speech, malicious advertisements, and pornography. To this end, content moderation software has been widely deployed on these platforms to detect and blocks toxic content. However, due to the complexity of content moderation models and the difficulty of understanding information across multiple modalities, existing content moderation software can fail to detect toxic content, which often leads to extremely negative impacts. We introduce Semantic Fusion, a general, effective methodology for validating multimedia content moderation software. Our key idea is to fuse two or more existing single-modal inputs (e.g., a textual sentence and an image) into a new input that combines the semantics of its ancestors in a novel manner and has toxic nature by construction. This fused input is then used for validating multimedia content moderation software. We realized Semantic Fusion as DUO, a practical content moderation software testing tool. In our evaluation, we employ DUO to test five commercial content moderation software and two state-of-the-art models against three kinds of toxic content. The results show that DUO achieves up to 100% error finding rate (EFR) when testing moderation software. In addition, we leverage the test cases generated by DUO to retrain the two models we explored, which largely improves model robustness while maintaining the accuracy on the original test set.

Yuting Yang, Binbin Du, Yingxin Zhang et al.

This paper introduces the system submitted by the Yidun NISP team to the video keyword wakeup challenge. We propose a mandarin keyword spotting system (KWS) with several novel and effective improvements, including a big backbone (B) model, a keyword biasing (B) mechanism and the introduction of syllable modeling units (S). By considering this, we term the total system BBS-KWS as an abbreviation. The BBS-KWS system consists of an end-to-end automatic speech recognition (ASR) module and a KWS module. The ASR module converts speech features to text representations, which applies a big backbone network to the acoustic model and takes syllable modeling units into consideration as well. In addition, the keyword biasing mechanism is used to improve the recall rate of keywords in the ASR inference stage. The KWS module applies multiple criteria to determine the absence or presence of the keywords, such as multi-stage matching, fuzzy matching, and connectionist temporal classification (CTC) prefix score. To further improve our system, we conduct semi-supervised learning on the CN-Celeb dataset for better generalization. In the VKW task, the BBS-KWS system achieves significant gains over the baseline and won the first place in two tracks.

Shuo Wang, Zhaopeng Tu, Zhixing Tan et al.

Recent years have witnessed the rapid advance in neural machine translation (NMT), the core of which lies in the encoder-decoder architecture. Inspired by the recent progress of large-scale pre-trained language models on machine translation in a limited scenario, we firstly demonstrate that a single language model (LM4MT) can achieve comparable performance with strong encoder-decoder NMT models on standard machine translation benchmarks, using the same training data and similar amount of model parameters. LM4MT can also easily utilize source-side texts as additional supervision. Though modeling the source- and target-language texts with the same mechanism, LM4MT can provide unified representations for both source and target sentences, which can better transfer knowledge across languages. Extensive experiments on pivot-based and zero-shot translation tasks show that LM4MT can outperform the encoder-decoder NMT model by a large margin.

Dabiao Ma, Zhiba Su, Wenxuan Wang et al.

End-to-end Text-to-speech (TTS) system can greatly improve the quality of synthesised speech. But it usually suffers form high time latency due to its auto-regressive structure. And the synthesised speech may also suffer from some error modes, e.g. repeated words, mispronunciations, and skipped words. In this paper, we propose a novel non-autoregressive, fully parallel end-to-end TTS system (FPETS). It utilizes a new alignment model and the recently proposed U-shape convolutional structure, UFANS. Different from RNN, UFANS can capture long term information in a fully parallel manner. Trainable position encoding and two-step training strategy are used for learning better alignments. Experimental results show FPETS utilizes the power of parallel computation and reaches a significant speed up of inference compared with state-of-the-art end-to-end TTS systems. More specifically, FPETS is 600X faster than Tacotron2, 50X faster than DCTTS and 10X faster than Deep Voice3. And FPETS can generates audios with equal or better quality and fewer errors comparing with other system. As far as we know, FPETS is the first end-to-end TTS system which is fully parallel.