Cybersecurity interviews uncover critical security experiences, behaviors, and privacy concerns often not fully captured by conventional technical assessments. Consequently, identifying genuine self-disclosures is challenging because statements are context-dependent, categories overlap, and automated interpretations ma...
Aisvarya Adeseye, Adeyemi Adeseye, Mohammad Tahir et al.· 0 citations
Spoken Dialogue Models (SDMs) typically share a speech-token space for generation and synthesis, leading to long autoregressive sequences and costly inference. Lower-rate speech tokenizers shorten sequences but may introduce compression-quality trade-offs and costly replacement without breaking this coupling. To addres...
Ying-Zhuo Deng, Chong Li, Bing Liu et al.· 0 citations
Large language models (LLMs) are increasingly used upstream of decision-making to clean data, detect label errors, and reconstruct missing values. Yet LLM-assisted data curation creates a new operational tradeoff: higher-fidelity models and richer prompts can be substantially more expensive, while the value of this add...
LLM-agent evaluation reports behavioral scores but leaves the identity of the deployed runtime stack unreported. A tampered deployment that preserves evaluated outputs passes every output-only evaluator. We introduce DIB-Bench, a benchmark that measures this deployment-integrity blindness across 7,560 scored examples,...
Jian-Chang Su, Da-Wei Xiang, Yi-Fan Zhang et al.· 0 citations
Label-free evaluation methods, such as self-consistency filtering, majority-vote aggregation, and agreement-based reliability scoring, rely on an implicit assumption that consistency can serve as a proxy for correctness when ground-truth labels are unavailable or expensive to obtain. In this paper, we test this assumpt...
Series decomposition has emerged as a widely acknowledged paradigm in deep time series forecasting models. However, existing approaches predominantly employ a homogeneous modeling strategy for all subsequences, thereby overlooking the distinct inductive biases inherent in different components. We identify that the tren...
Large-scale multi-agent ecosystems require mechanisms for discovering suitable agents without exposing entire registries to a large language model (LLM) orchestrator. The challenge is not only to retrieve relevant Agent Cards from a growing registry, but also to maintain a compact candidate context, since each addition...
Giancarlo Maldonado C\'ardenas, Carlos A. Astudillo, Nelson L. S. da Fonseca· 0 citations
The success score of a tool agent reflects both its model and the budgets of the harness that runs it. We present ToolBottleneckBench, a benchmark that measures tool-agent performance under controlled pressures and fixed harness budgets. Synthetic enterprise tasks stay fixed while one pressure varies: the registered to...
Zichuan Xu, Yuyang Dong, Makoto Onizuka et al.· 0 citations
We identify a theoretical incompatibility between 3D point cloud masked autoencoding and SE(3)-equivariant networks. We prove that standard masking triggers a topological collapse, effectively forcing the model to hallucinate orientation without a reference frame. To resolve this, we introduce SIEVE (Scalar Invariant E...
Jongann Lee, Sun Woo Park, Yun Young Choi· 0 citations
The jury, with nine experts of up to 3B parameters, performs comparably to the same fusion rule with label-fitted weights, showing no statistically significant advantage from supervision, and outperforms the best single model with matched self-consistency and the strongest label-free alternative on the full benchmark.
A. Rozza, D. Mambelli, S. Motta et al.· 0 citations
AI agents now pause for approval before they send money, delete files or email outsiders, and that approval is usually treated as a second factor. We argue it is one only when compromising the agent cannot produce it and when it names exactly what it approves, and we test both conditions. We define twelve attacks on th...
Sai Varun Kodathala, Prashanth Pollishetty, J. Cargill· 2 citations
