• 学术搜索
  • 科研智能体
    • Research Labs
    • AI 阅读
    • AI 文库
    • 深度研究
    • 学者亮点
  • 学术资源
    • AI2000
    • 期刊/会议
    • 学者库
    • 学术API
      MCP&Skill
    • 溯源树
    • 数据集
  • 知识沉淀
    • 学术空间
订阅小程序
旧版功能
aminer vip
开通会员低至0.73元/天
一次搞定AI科研
立即登录
  • English
  • 联系方式
    A

    American Physical Therapy Association

    EST. 1921
    1,520论文总数
    4.2万引用总数

    论文量&引用量时间轴

    机构学者

    论文(1520)

    年份
    起
    –
    止
    排序
    1Hidden in Memory: Sleeper Memory Poisoning in LLM Agents
    Sidharth Pulipaka, Stanislau Hlebik, Leonidas Raghav,Sahar Abdelnabi,Vyas Raina,Ivaxi Sheth,Mario Fritz

    Large language models are increasingly augmented with persistent memory, allowing assistants to store user-specific information across sessions for personalization and continuity. This statefulness introduces a new security risk: adversarial content can corrupt what an assistant remembers and thereby influence future interactions. We propose and study sleeper memory poisoning, a delayed attack in which an adversary manipulates external context, such as a document, webpage, or repository, to cause the assistant to store a fabricated memory about the user. Unlike conventional prompt injection, the attack can remain dormant and re-emerge across multiple later conversations. We evaluate the full attack pipeline: whether poisoned memories are written, later retrieved, and ultimately used to steer the following conversations. Across stateful LLM assistants, poisoned memories were added up to 99.8

    2026引用:33
    引用
    AI阅读
    加入学术空间
    2Funny or Persuasive, but Not Both: Evaluating Fine-Grained Multi-Concept Control in LLMs
    Arya Labroo,Ivaxi Sheth,Vyas Raina, Amaani Ahmed,Mario Fritz

    Large Language Models (LLMs) offer strong generative capabilities, but many applications require explicit and fine-grained control over specific textual concepts, such as humor, persuasiveness, or formality. Prior approaches in prompting and representation engineering can provide coarse or single-attribute control, but systematic evaluation of multi-attribute settings remains limited. We introduce an evaluation framework for fine-grained controllability for both single- and dual-concept scenarios, focusing on linguistically distinct concept pairs (e.g., persuasiveness vs. humor). Surprisingly, across multiple LLMs and generative tasks, we find that performance often drops in the dual-concept setting, even though the chosen concepts should in principle be separable. This reveals a fundamental limitation of naive prompting-based control: models struggle with compositionality even when concepts are intuitively independent. Our framework provides systematic evidence of this gap and offers a principled approach for measuring the ability of future methods for multi-concept control.

    2026PROCEEDINGS OF THE 19TH CONFERENCE OF THE EUROPEAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LING...(2026)引用:2
    引用
    AI阅读
    加入学术空间
    32026 Guideline for Adult Stroke Rehabilitation and Recovery: A Guideline from the American Heart Association and American Stroke Association.
    Lorie G Richards,Nneka L Ifejika,Joel Stein, Mona N Bahouth, Martha Abshire Saylor, A M Barrett,Peii Chen,Anne F Deutsch,Wendy Dusenbury,Charles Ellis,Janice Eng, Jamie Fleet,

    AIM:The "2026 Guidelines for Adult Stroke Rehabilitation and Recovery" replaces the 2016 "Guidelines for Adult Stroke Rehabilitation and Recovery." This updated guideline is intended to provide a comprehensive, up-to-date, evidence-based set of recommendations. The intended audience includes physicians, allied health professionals, and caregivers. METHODS:A comprehensive search for literature published since the 2016 guideline, derived from research involving human participants, published in English, and indexed in MEDLINE, PubMed, Cochrane Library, and other selected databases relevant to this guideline, was conducted between June and October 2025. Other documents on related subject matter previously published by the American Heart Association were also reviewed. STRUCTURE:This guideline provides recommendations based on the most current evidence available for stroke rehabilitation and recovery. Key updates include new and revised recommendations across the stroke recovery continuum, including the assessment and treatment of medical comorbidities, fracture risk, fall prevention, technology-enabled rehabilitation, caregiver support, participation, and safe return to work and driving. Although this guideline reflects significant advances, it also highlights gaps in knowledge and underscores the urgent need for continued research to further refine and improve treatment strategies.

    2026Stroke(2026)引用:1
    引用
    AI阅读
    加入学术空间
    4Share-Borne AI Virus: Memory-Hopping Attacks Across LLM Agents
    Sidharth Pulipaka, Ansh Sharma, Stanislau Hlebik, Leonidas Raghav,Vyas Raina,Ivaxi Sheth, Mario Fritz

    Large language models are increasingly deployed as stateful assistants that retain information across interactions and use tools to read, modify, and create persistent artifacts. As these artifacts are shared between users, they form an indirect communication channel between otherwise independent assistants. We study a failure mode in which this channel enables self-propagating attacks. We introduce artifact-mediated propagation, where adversarial content introduced through an artifact (e.g. a report), is stored in an assistant's persistent memory, reproduced in a subsequently created artifact, and acquired by another assistant that later reads it. We evaluate this process in temporal human-agent universes that model artifact exchange between independently operated assistants over time, measuring whether an attack survives successive hand-offs, how many hops it reaches, and how broadly it spreads. We find that attacks can propagate across multiple independent assistants and persist over extended interaction sequences. In larger simulated environments, even GPT-5.6 Luna exhibits substantial spread, reaching 60-80

    2026引用:1
    引用
    AI阅读
    加入学术空间
    5Soft Contextualized Encoder for User Defined Text Classification
    Charu Maheshwari, Vyas Raina

    User-Defined Text Classification (UDTC) considers the challenge of classifying input text to user-specified, previously unseen classes, a setting that arises frequently in real-world applications such as enterprise analytics, content moderation, and domain-specific information retrieval. We propose a soft-contextualized encoder architecture for UDTC which contextualizes each candidate label with the label set and a static soft prompt representation of the input query. Training on diverse, multi-source datasets enables the model to generalize effectively to zero-shot classification over entirely unseen topic sets drawn from arbitrary domains. We evaluate the proposed architecture both on held-out in-distribution test data and on multiple unseen UDTC benchmarks. Across datasets, the model achieves state-of-the-art performance, consistently outperforming or matching the baselines.

    2026CoRR(2026)
    引用
    AI阅读
    加入学术空间
    立即登录,查看全部 1520 篇论文

    合作机构(100)

    圣保罗州立大学合作论文 67
    圣保罗大学合作论文 27
    捷克科学院合作论文 22
    坎皮纳斯州立大学合作论文 22
    开罗大学合作论文 22
    巴西农业研究公司合作论文 17
    Agência Paulista de Tecnologia dos Agronegócios合作论文 14
    东京大学合作论文 13
    密歇根大学合作论文 11
    内布拉斯加大学医学中心合作论文 9

    机构统计