~ similar to 2604.03612v1· 17 results
This paper proposes and implements the first Bengali text CAPTCHA mechanism for native Bengali-speaking users, achieving high security and usability.
This study investigates how humans detect synthetic speech in real-world contexts, finding that while overt detection failed for fully synthetic speech, participants still implicitly discriminated utt…
Xinhao Song, Su Su, Sirui Song, Hongliang Wu +5 more
The paper introduces HLL, a benchmark that tests if multimodal agents can successfully substitute for human verification (like CAPTCHA) in complex, real-world workflows, finding that current agents ar…
The paper introduces DECKER, a domain-invariant framework that significantly improves cross-keyboard keystroke inference by normalizing device variations and leveraging linguistic context, demonstrati…
Yuxi Chen, Haoyu Zhai, Chenkai Wang, Rui Yang +3 more
The paper introduces ReCAP, a native GUI agent that significantly improves CAPTCHA solving success (from 30% to 80%) by integrating specialized CAPTCHA capabilities into a general-purpose, end-to-end…
This paper evaluates the use of large audio language models for speaker verification systems against conventional pipelines and finds that task-specific adaptation improves performance.
The paper introduces MusICA-MetaBench, a framework for deriving on-demand music perception benchmarks from user-provided data, ensuring statistically reliable model comparisons.
Aniket Anand, Janvijay Singh, Zhewei Sun, Dilek Hakkani-Tür +1 more
The paper demonstrates that the AI-like style introduced by post-training alignment can be measured, localized, and causally removed using a novel ablation technique called PASTA.
This survey provides a comprehensive taxonomy and vulnerability-centric analysis of adversarial attacks targeting Multimodal Large Language Models (MLLMs), offering an explanatory framework for enhanc…
Yifan Liao, Zongmin Zhang, Zhen Sun, Yuhui Sun +2 more
The paper introduces a novel Clean-Referenced Feature-Vocoder Attack, a black-box adversarial attack that perturbs high-level SSL feature representations instead of raw audio waveforms, achieving supe…
This paper evaluates the effectiveness of interactive and non-interactive bot defense systems against commercial Captcha-solving services and LLM-based browser agents.
Alexander Nemecek, Osama Zafar, Yuqiao Xu, Wenbiao Li +1 more
The paper argues that current AI content watermarking benchmarks fail to test for bias across different languages, cultures, and demographics, proposing a new set of evaluation standards to ensure fai…
This paper demonstrates that typographic attacks pose a significant, measurable, and physically consequential threat to household robot manipulation systems by causing the robot to grasp and transport…
Meng Chen, Kun Wang, Li Lu, Jiaheng Zhang +1 more
The paper introduces AudioHijack, a framework that successfully demonstrates context-agnostic and imperceptible auditory prompt injection attacks, showing that commercial Large Audio-Language Models c…
Sicheng Yang, Shulan Ruan, Shiwei Wu, Yu Liu +3 more
PolySpeech-100 introduces a massive, multi-lingual benchmark covering 110 linguistic variants to rigorously test Speech-LLMs, demonstrating that open-source models struggle with low-resource languages…
The paper investigates anthropomorphic reflection markers (like 'hmm' or 'wait') in LLM reasoning and finds that these markers are often surface cues, not necessary for strong reasoning performance.
This paper compares the performance of speech deepfake countermeasures using equal error rate (EER) and half total error rate (HTER) on different datasets. It also evaluates the effectiveness of popul…