arXiv subjects

Yuhan Liu

Publications and source records attributed to Yuhan Liu.

At least 109 records · Page 6Linked to original sources

arXiv2025

Trading Mathematical for Physical Simplicity: Bialgebraic Structures in Matrix Product Operator Symmetries

Despite recent advances in the lattice representation theory of (generalized) symmetries, many simple quantum spin chains of physical interest are not included in the rigid framework of fusion categories and weak Hopf algebras. We demonstrate that this problem can be overcome by relaxing the requirements on the underlying algebraic structure, and show that general matrix product operator symmetries are described by a pre-bialgebra. As a guiding example, we focus on the anomalous $\mathbb Z_2$ symmetry of the XX model, which manifests the mixed anomaly between its $U(1)$ momentum and winding symmetry. We show how this anomaly is embedded into the non-semisimple corepresentation category, providing a novel mechanism for realizing such anomalous symmetries on the lattice. Additionally, the representation category which describes the renormalization properties is semisimple and semi-monoidal, which provides a new class of mixed state renormalization fixed points. Finally, we show that up to a quantum channel, this anomalous $\mathbb Z_2$ symmetry is equivalent to a more conventional MPO symmetry obtained on the boundary of a double semion model. In this way, our work provides a bridge between well-understood topological defect symmetries and those that arise in more realistic models.

quant-ph↗

arXiv2025

Injecting Domain-Specific Knowledge into Large Language Models: A Comprehensive Survey

Large Language Models (LLMs) have demonstrated remarkable success in various tasks such as natural language understanding, text summarization, and machine translation. However, their general-purpose nature often limits their effectiveness in domain-specific applications that require specialized knowledge, such as healthcare, chemistry, or legal analysis. To address this, researchers have explored diverse methods to enhance LLMs by integrating domain-specific knowledge. In this survey, we provide a comprehensive overview of these methods, which we categorize into four key approaches: dynamic knowledge injection, static knowledge embedding, modular adapters, and prompt optimization. Each approach offers unique mechanisms to equip LLMs with domain expertise, balancing trade-offs between flexibility, scalability, and efficiency. We discuss how these methods enable LLMs to tackle specialized tasks, compare their advantages and disadvantages, evaluate domain-specific LLMs against general LLMs, and highlight the challenges and opportunities in this emerging field. For those interested in delving deeper into this area, we also summarize the commonly used datasets and benchmarks. To keep researchers updated on the latest studies, we maintain an open-source at: https://github.com/abilliyb/Knowledge_Injection_Survey_Papers, dedicated to documenting research in the field of specialized LLM.

cs.CL↗

arXiv2025

Adversarially robust quantum state learning and testing

Quantum state learning is a fundamental problem in physics and computer science. As near-term quantum devices are error-prone, it is important to design error-resistant algorithms. Apart from device errors, other unexpected factors could also affect the algorithm, such as careless human read-out error, or even a malicious hacker deliberately altering the measurement results. Thus, we want our algorithm to work even in the worst case when things go against our favor. We consider the practical setting of single-copy measurements and propose the $γ$-adversarial corruption model where an imaginary adversary can arbitrarily change $γ$-fraction of the measurement outcomes. This is stronger than the $γ$-bounded SPAM noise model, where the post-measurement state changes by at most $γ$ in trace distance. Under our stronger model of corruption, we design an algorithm using non-adaptive measurements that can learn an unknown rank-$r$ state up to $\tilde{O}(γ\sqrt{r})$ in trace distance, provided that the number of copies is sufficiently large. We further prove an information-theoretic lower bound of $Ω(γ\sqrt{r})$ for non-adaptive measurements, demonstrating the optimality of our algorithm. Our upper and lower bounds also hold for quantum state testing, where the goal is to test whether an unknown state is equal to a given state or far from it. Our results are intriguingly optimistic and pessimistic at the same time. For general states, the error is dimension-dependent and $γ\sqrt{d}$ in the worst case, meaning that only corrupting a very small fraction ($1/\sqrt{d}$) of the outcomes could totally destroy any non-adaptive learning algorithm. However, for constant-rank states that are useful in many quantum algorithms, it is possible to achieve dimension-independent error, even in the worst-case adversarial setting.

quant-ph↗

arXiv2025

CEM-Net: Cross-Emotion Memory Network for Emotional Talking Face Generation

Emotional talking face generation aims to animate a human face in given reference images and generate a talking video that matches the content and emotion of driving audio. However, existing methods neglect that reference images may have a strong emotion that conflicts with the audio emotion, leading to severe emotion inaccuracy and distorted generated results. To tackle the issue, we introduce a cross-emotion memory network(CEM-Net), designed to generate emotional talking faces aligned with the driving audio when reference images exhibit strong emotion. Specifically, an Audio Emotion Enhancement module(AEE) is first devised with the cross-reconstruction training strategy to enhance audio emotion, overcoming the disruption from reference image emotion. Secondly, since reference images cannot provide sufficient facial motion information of the speaker under audio emotion, an Emotion Bridging Memory module(EBM) is utilized to compensate for the lacked information. It brings in expression displacement from the reference image emotion to the audio emotion and stores it in the memory.Given a cross-emotion feature as a query, the matching displacement can be retrieved at inference time. Extensive experiments have demonstrated that our CEM-Net can synthesize expressive, natural and lip-synced talking face videos with better emotion accuracy.

cs.MM↗

arXiv2025

HarmonyGuard: Toward Safety and Utility in Web Agents via Adaptive Policy Enhancement and Dual-Objective Optimization

Large language models enable agents to autonomously perform tasks in open web environments. However, as hidden threats within the web evolve, web agents face the challenge of balancing task performance with emerging risks during long-sequence operations. Although this challenge is critical, current research remains limited to single-objective optimization or single-turn scenarios, lacking the capability for collaborative optimization of both safety and utility in web environments. To address this gap, we propose HarmonyGuard, a multi-agent collaborative framework that leverages policy enhancement and objective optimization to jointly improve both utility and safety. HarmonyGuard features a multi-agent architecture characterized by two fundamental capabilities: (1) Adaptive Policy Enhancement: We introduce the Policy Agent within HarmonyGuard, which automatically extracts and maintains structured security policies from unstructured external documents, while continuously updating policies in response to evolving threats. (2) Dual-Objective Optimization: Based on the dual objectives of safety and utility, the Utility Agent integrated within HarmonyGuard performs the Markovian real-time reasoning to evaluate the objectives and utilizes metacognitive capabilities for their optimization. Extensive evaluations on multiple benchmarks show that HarmonyGuard improves policy compliance by up to 38% and task completion by up to 20% over existing baselines, while achieving over 90% policy compliance across all tasks. Our project is available here: https://github.com/YurunChen/HarmonyGuard.

cs.CL↗

arXiv2025

Pauli Measurements Are Near-Optimal for Single-Qubit Tomography

We provide the first non-trivial lower bounds for single-qubit tomography algorithms and show that at least $Ω\left(\frac{10^N}{\sqrt{N} \varepsilon^2}\right)$ copies are required to learn an $N$-qubit state $ρ\in\mathbb{C}^{d\times d},d=2^N$ to within $\varepsilon$ trace distance. Pauli measurements, the most commonly used single-qubit measurement scheme, have recently been shown to require at most $O\left(\frac{10^N}{\varepsilon^2}\right)$ copies for this problem. Combining these results, we nearly settle the long-standing question of the complexity of single-qubit tomography.

quant-ph↗

arXiv2025

DroidSpeak: KV Cache Sharing for Cross-LLM Communication and Multi-LLM Serving

Compound AI systems, such as agentic systems, are an emerging trend in large-scale enterprise settings, with multiple LLMs specialized for different users, tasks, and/or roles working together. In these scenarios, different models often process inputs that share the same context prefix. Although much work was done in the past to enable the reuse of prefix KV caches across inputs for a single model, how to enable one model to reuse the prefix KV caches of a different model remains an open question. We introduce DroidSpeak, the first distributed LLM inference system that enables KV cache reuse across distributed nodes running inference of different LLMs, so long as the LLMs have the same architecture. We present the first study that aims at understanding the impact of sharing KV caches across different LLMs, and if/when such sharing affects quality. Inspired by the findings, we present DroidSpeak, which selectively recomputes a few layers of the KV cache produced by another LLM and reuses the remaining layers, with negligible quality loss. Moreover, carefully pipelining the layer-wise re-computation and the loading of reused KV cache further improves the inference performance. Experiments on diverse datasets and model pairs demonstrate that DroidSpeak achieves up to 4x throughput improvement and about 3.1x faster prefill (time to first token), with negligible loss of quality in F1 scores, Rouge-L or code similarity score, compared to the baseline which does not allow any sharing across models.

cs.MA↗

arXiv2025

Mind the Gap: Aligning Vision Foundation Models to Image Feature Matching

Leveraging the vision foundation models has emerged as a mainstream paradigm that improves the performance of image feature matching. However, previous works have ignored the misalignment when introducing the foundation models into feature matching. The misalignment arises from the discrepancy between the foundation models focusing on single-image understanding and the cross-image understanding requirement of feature matching. Specifically, 1) the embeddings derived from commonly used foundation models exhibit discrepancies with the optimal embeddings required for feature matching; 2) lacking an effective mechanism to leverage the single-image understanding ability into cross-image understanding. A significant consequence of the misalignment is they struggle when addressing multi-instance feature matching problems. To address this, we introduce a simple but effective framework, called IMD (Image feature Matching with a pre-trained Diffusion model) with two parts: 1) Unlike the dominant solutions employing contrastive-learning based foundation models that emphasize global semantics, we integrate the generative-based diffusion models to effectively capture instance-level details. 2) We leverage the prompt mechanism in generative model as a natural tunnel, propose a novel cross-image interaction prompting module to facilitate bidirectional information interaction between image pairs. To more accurately measure the misalignment, we propose a new benchmark called IMIM, which focuses on multi-instance scenarios. Our proposed IMD establishes a new state-of-the-art in commonly evaluated benchmarks, and the superior improvement 12% in IMIM indicates our method efficiently mitigates the misalignment.

cs.CV↗

arXiv2025

Failure Forecasting Boosts Robustness of Sim2Real Rhythmic Insertion Policies

This paper addresses the challenges of Rhythmic Insertion Tasks (RIT), where a robot must repeatedly perform high-precision insertions, such as screwing a nut into a bolt with a wrench. The inherent difficulty of RIT lies in achieving millimeter-level accuracy and maintaining consistent performance over multiple repetitions, particularly when factors like nut rotation and friction introduce additional complexity. We propose a sim-to-real framework that integrates a reinforcement learning-based insertion policy with a failure forecasting module. By representing the wrench's pose in the nut's coordinate frame rather than the robot's frame, our approach significantly enhances sim-to-real transferability. The insertion policy, trained in simulation, leverages real-time 6D pose tracking to execute precise alignment, insertion, and rotation maneuvers. Simultaneously, a neural network predicts potential execution failures, triggering a simple recovery mechanism that lifts the wrench and retries the insertion. Extensive experiments in both simulated and real-world environments demonstrate that our method not only achieves a high one-time success rate but also robustly maintains performance over long-horizon repetitive tasks.

cs.RO↗

arXiv2025

Divide-Fuse-Conquer: Eliciting "Aha Moments" in Multi-Scenario Games

Large language models (LLMs) have been observed to suddenly exhibit advanced reasoning abilities during reinforcement learning (RL), resembling an ``aha moment'' triggered by simple outcome-based rewards. While RL has proven effective in eliciting such breakthroughs in tasks involving mathematics, coding, and vision, it faces significant challenges in multi-scenario games. The diversity of game rules, interaction modes, and environmental complexities often leads to policies that perform well in one scenario but fail to generalize to others. Simply combining multiple scenarios during training introduces additional challenges, such as training instability and poor performance. To overcome these challenges, we propose Divide-Fuse-Conquer, a framework designed to enhance generalization in multi-scenario RL. This approach starts by heuristically grouping games based on characteristics such as rules and difficulties. Specialized models are then trained for each group to excel at games in the group is what we refer to as the divide step. Next, we fuse model parameters from different groups as a new model, and continue training it for multiple groups, until the scenarios in all groups are conquered. Experiments across 18 TextArena games show that Qwen2.5-32B-Align trained with the Divide-Fuse-Conquer strategy reaches a performance level comparable to Claude3.5, achieving 7 wins and 4 draws. We hope our approach can inspire future research on using reinforcement learning to improve the generalization of LLMs.

cs.LG↗

arXiv2025

Understanding Community-Level Blocklists in Decentralized Social Media

Community-level blocklists are key to content moderation practices in decentralized social media. These blocklists enable moderators to prevent other communities, such as those acting in bad faith, from interacting with their own -- and, if shared publicly, warn others about communities worth blocking. Prior work has examined blocklists in centralized social media, noting their potential for collective moderation outcomes, but has focused on blocklists as individual-level tools. To understand how moderators perceive and utilize community-level blocklists and what additional support they may need, we examine social media communities running Mastodon, an open-source microblogging software built on the ActivityPub protocol. We conducted (1) content analysis of the community-level blocklist ecosystem, and (2) semi-structured interviews with twelve Mastodon moderators. Our content analysis revealed wide variation in blocklist goals, inclusion criteria, and transparency. Interviews showed moderators balance proactive safety, reactive practices, and caution around false positives when using blocklists for moderation. They noted challenges and limitations in current blocklist use, suggesting design improvements like comment receipts, category filters, and collaborative voting. We discuss implications for decentralized content moderation, highlighting trade-offs between openness, safety, and nuance; the complexity of moderator roles; and opportunities for future design.

cs.SI↗

arXiv2025

GPTFootprint: Increasing Consumer Awareness of the Environmental Impacts of LLMs

With the growth of AI, researchers are studying how to mitigate its environmental impact, primarily by proposing policy changes and increasing awareness among developers. However, research on AI end users is limited. Therefore, we introduce GPTFootprint, a browser extension that aims to increase consumer awareness of the significant water and energy consumption of LLMs, and reduce unnecessary LLM usage. GPTFootprint displays a dynamically updating visualization of the resources individual users consume through their ChatGPT queries. After a user reaches a set query limit, a popup prompts them to take a break from ChatGPT. In a week-long user study, we found that GPTFootprint increases people's awareness of environmental impact, but has limited success in decreasing ChatGPT usage. This research demonstrates the potential for individual-level interventions to contribute to the broader goal of sustainable AI usage, and provides insights into the effectiveness of awareness-based behavior modification strategies in the context of LLMs.

cs.HC↗

arXiv2025

ConversAR: Exploring Embodied LLM-Powered Group Conversations in Augmented Reality for Second Language Learners

Group conversations are valuable for second language (L2) learners as they provide opportunities to practice listening and speaking, exercise complex turn-taking skills, and experience group social dynamics in a target language. However, most existing Augmented Reality (AR)-based conversational learning tools focus on dyadic interactions rather than group dialogues. Although research has shown that AR can help reduce speaking anxiety and create a comfortable space for practicing speaking skills in dyadic scenarios, especially with Large Language Model (LLM)-based conversational agents, the potential for group language practice using these technologies remains largely unexplored. We introduce ConversAR, a gpt-4o powered AR application, that enables L2 learners to practice contextualized group conversations. Our system features two embodied LLM agents with vision-based scene understanding and live captions. In a system evaluation with 10 participants, users reported reduced speaking anxiety and increased learner autonomy compared to perceptions of in-person practice methods with other learners.

cs.HC↗

arXiv2025

PolicyPulse: LLM-Synthesis Tool for Policy Researchers

Public opinion shapes policy, yet capturing it effectively to surface diverse perspectives remains challenging. This paper introduces PolicyPulse, an LLM-powered interactive system that synthesizes public experiences from online community discussions to help policy researchers author memos and briefs, leveraging curated real-world anecdotes. Given a specific topic (e.g., "Climate Change"), PolicyPulse returns an organized list of themes (e.g., "Biodiversity Loss" or "Carbon Pricing"), supporting each theme with relevant quotes from real-life anecdotes. We compared PolicyPulse outputs to authoritative policy reports. Additionally, we asked 11 policy researchers across multiple institutions in the Northeastern U.S to compare using PolicyPulse with their expert approach. We found that PolicyPulse's themes aligned with authoritative reports and helped spark research by analyzing existing data, gathering diverse experiences, revealing unexpected themes, and informing survey or interview design. Participants also highlighted limitations including insufficient demographic context and data verification challenges. Our work demonstrates how AI-powered tools can help influence policy-relevant research and shape policy outcomes.

cs.HC↗

arXiv2025

The Stepwise Deception: Simulating the Evolution from True News to Fake News with LLM Agents

With the growing spread of misinformation online, understanding how true news evolves into fake news has become crucial for early detection and prevention. However, previous research has often assumed fake news inherently exists rather than exploring its gradual formation. To address this gap, we propose FUSE (Fake news evolUtion Simulation framEwork), a novel Large Language Model (LLM)-based simulation approach explicitly focusing on fake news evolution from real news. Our framework model a social network with four distinct types of LLM agents commonly observed in daily interactions: spreaders who propagate information, commentators who provide interpretations, verifiers who fact-check, and bystanders who observe passively to simulate realistic daily interactions that progressively distort true news. To quantify these gradual distortions, we develop FUSE-EVAL, a comprehensive evaluation framework measuring truth deviation along multiple linguistic and semantic dimensions. Results show that FUSE effectively captures fake news evolution patterns and accurately reproduces known fake news, aligning closely with human evaluations. Experiments demonstrate that FUSE accurately reproduces known fake news evolution scenarios, aligns closely with human judgment, and highlights the importance of timely intervention at early stages. Our framework is extensible, enabling future research on broader scenarios of fake news.

cs.SI↗

arXiv2025

SAGraph: A Large-Scale Social Graph Dataset with Comprehensive Context for Influencer Selection in Marketing

Influencer marketing campaign success heavily depends on identifying key opinion leaders who can effectively leverage their credibility and reach to promote products or services. The selecting influencers process is vital for boosting brand visibility, fostering consumer trust, and driving sales. While traditional research often simplifies complex factors like user attitudes, interaction frequency, and advertising content, into simple numerical values. However, this reductionist approach fails to capture the dynamic nature of influencer marketing effectiveness. To bridge this gap, we present SAGraph, a novel comprehensive dataset from Weibo that captures multi-dimensional marketing campaign data across six product domains. The dataset encompasses 345,039 user profiles with their complete interaction histories, including 1.3M comments and 554K reposts across 44K posts, providing unprecedented granularity in influencer marketing dynamics. SAGraph uniquely integrates user profiles, content features, and temporal interaction patterns, enabling in-depth analysis of influencer marketing mechanisms. Experimental results using both traditional baselines and state-of-the-art large language models (LLMs) demonstrate the crucial role of content analysis in predicting advertising effectiveness. Our findings reveal that LLM-based approaches achieve superior performance in understanding and predicting campaign success, opening new avenues for data-driven influencer marketing strategies. We hope that this dataset will inspire further research https://github.com/xiaoqzhwhu/SAGraph/.

cs.SI↗

arXiv2025

More is not always better? Enhancing Many-Shot In-Context Learning with Differentiated and Reweighting Objectives

Large language models (LLMs) excel at few-shot in-context learning (ICL) without requiring parameter updates. However, as ICL demonstrations increase from a few to many, performance tends to plateau and eventually decline. We identify two primary causes for this trend: the suboptimal negative log-likelihood (NLL) optimization objective and the incremental data noise. To address these issues, we introduce \textit{DrICL}, a novel optimization method that enhances model performance through \textit{Differentiated} and \textit{Reweighting} objectives. Globally, DrICL utilizes differentiated learning to optimize the NLL objective, ensuring that many-shot performance surpasses zero-shot levels. Locally, it dynamically adjusts the weighting of many-shot demonstrations by leveraging cumulative advantages inspired by reinforcement learning, thereby mitigating the impact of noisy data. Recognizing the lack of multi-task datasets with diverse many-shot distributions, we develop the \textit{Many-Shot ICL Benchmark} (ICL-50)-a large-scale benchmark of 50 tasks that cover shot numbers from 1 to 350 within sequences of up to 8,000 tokens-for both fine-tuning and evaluation purposes. Experimental results demonstrate that LLMs enhanced with DrICL achieve significant improvements in many-shot setups across various tasks, including both in-domain and out-of-domain scenarios. We release the code and dataset hoping to facilitate further research in many-shot ICL\footnote{https://github.com/xiaoqzhwhu/DrICL}.

cs.LG↗

arXiv2025

Thinking Before Running! Efficient Code Generation with Thorough Exploration and Optimal Refinement

Code generation is crucial in software engineering for automating the coding process efficiently. While test-time computation methods show promise, they suffer from high latency due to multiple computation rounds. To overcome this, we introduce \textbf{ThinkCoder}, a framework that combines thorough exploration with optimal refinement. The exploration phase diversifies the solution space by searching for potential solutions, followed by a refinement phase that enhances precision. This approach allows us to select the best solution through careful consideration before taking action, avoiding excessive trial and error. To further minimize test-time computation overhead, we introduce preference-driven optimization with Reinforced Self-Training (ReST), which uses exploration trajectories from ThinkCoder to guide LLM's evolution. This approach enhances LLM's exploration efficiency via preference learning, cutting costs while maintaining accuracy. ThinkCoder boosts the performance with a single LLM, excelling on benchmarks like HumanEval and MBPP. Compared to SOTA models, it improves Pass@1 by 3.0\% over MapCoder with just 6.4\% of the computation cost. Against AgentCoder, ThinkCoder achieves a 0.5\% higher Pass@1 after 2 rounds, outperforming AgentCoder's 5 rounds. Additionally, ReST with success trajectories enhances efficiency, allowing models like LLaMA2-7B to achieve competitive results using only 20\% of the computational resources. These results highlight the framework's effectiveness and scalability.

cs.SE↗