Extracting MITRE ATT&CK techniques from cyber threat intelligence (CTI) reports is an open-set, multi-label problem requiring both high recall (not missing techniques) and high precision (not hallucinating unsupported ones). Existing rule-based, supervised, and LLM-based methods struggle to achieve both: rule-based and...
Yutong Cheng, Changze Li, Raihan Sultan Pasha Basuki et al.· 0 citations
Third-party Python libraries introduce dependency-management overhead, supply-chain risk, and deployment friction. We study when eliminating a dependency preserves enough correctness and performance to justify the gains in footprint, deployability, and trust surface. Our testbed is zerodep, 44 single-file, stdlib-only...
Symmetry determines which physical responses of molecules and crystals are allowed and underlies many models used throughout the physical sciences. Yet several widely used machine-learned models learn symmetry only approximately from data, raising the question of whether approximate symmetry is sufficient for scientifi...
Paulina Hoyos, Shashanka Ubaru, Dongsung Huh et al.· 0 citations
Training data attribution (TDA) should enable generative model interpretability and foster a variety of related downstream tasks. Nonetheless, current TDA approaches lack reliability and robustness, preventing their adoption in real-world setups. In this paper, we take a decisive step towards more reliable and robust T...
Joan Serr\`a, Dipam Goswami, Fabio Morreale et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Scaling on-policy distillation (OPD) for large language models (LLMs) confronts a fundamental tension: asynchronous execution is necessary for system efficiency, but structurally deviates from the ideal on-policy objective. To address this challenge, we theoretically decompose the objective discrepancy into rollout dri...
Traffic sensor data exhibit strong temporal and cross-sensor dependencies, yet long gaps erase both a sensor's local history and the joint observations needed to learn its relationships with other sensors. We introduce MoHL, a mixture of hypergraph Laplacians that learns group-wise relational structure directly from in...
Automation can displace or complement labour, but this need not be constant across economies. Existing exposure measures typically assign fixed scores to tasks or occupations and capture cross-country variation through employment structure. We posit that exposure to automation depends jointly on task content and countr...
Search engines traditionally complement online content platforms by directing users seeking information to external websites. The emergence of generative AI search tools that summarize answers directly on the results page may disrupt this relationship by making visits to source platforms optional. We study this questio...
Peibo Zhang, Ruomeng Cui, Dennis J. Zhang· 0 citations
Looped language models (LoopLMs) increase effective computational depth by repeatedly applying shared Transformer blocks without increasing the parameter count. However, this reuse makes them particularly sensitive to post-training quantization (PTQ). We systematically study PTQ in LoopLMs and identify three sources of...
Rui Fang, Hsi-Wen Chen, Chia-Ming Cheng et al.· 0 citations
Multilingual speech translation frontends built on speech large language models typically condition on language identity, as a learned embedding or a prompted language name. Such a signal, however, is learned separately for each language and transfers nothing between typologically related languages. In this paper, we p...
Yu Pan, Tianjiao Zhang, Yang Hou et al.· 0 citations
Concept Activation Vectors (CAV) and the Testing with CAVs (TCAV) method are fundamental tools for concept-based explainability in deep learning, yet their practical utility is limited by the variability across different samples of concept and non-concept examples. We analyze this stochastic nature of CAVs, and identif...
Ekkehard Schnoor, Jawher Said, Malik Tiomoko et al.· 0 citations
Function calling, also known as tool use, is a core capability of modern LLM agents but is typically constrained by synchronous execution semantics. Under these semantics, LLM decoding is blocked until each function call completes, resulting in increasing end-to-end latency. In this work, we introduce AsyncFC, a pure e...
Guangyu Feng, Huanzhi Mao, Prabal Dutta et al.· 0 citations
