Generative recommendation models based on autoregressive decoding have been adopted as alternatives to classical ranking-based systems. While their modeling flexibility has been widely studied, their resilience to data poisoning remains largely unexplored. In this work, we ask: how susceptible are generative recommende...
Zhuo-Xuan Zhang, Jinhao Duan, Xin-Yu Zhao et al.· 0 citations
Enterprise workflow automation typically requires retrieving a complete and accurate set of relevant information from large operational document corpora (SOPs, policy documents) given a specific scenario or request. In customer support (CS), for example, human specialists must determine the correct sequence of resoluti...
Istok Bako, Bonan Min, Tiffany Shao et al.· 0 citations
Large language models (LLMs) have achieved strong performance in mathematical and logical reasoning, yet they still suffer from the curse of complexity and struggle as problem difficulty increases with the number of constraint interactions and the size of the search space. In contrast, satisfiability solvers such as Z3...
Benchmark scores tell us what an LLM can do, but not how much energy is required to deliver that capability. We directly measure this trade-off for 86 open-weight model checkpoints, evaluated on seven benchmarks and six hardware platforms ranging from a consumer GPU to an eight-GPU server. We find large differences in...
Ze-Yu Yang, Wesley Armour· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Agentic AI systems operating over geospatial data require more than autonomous reasoning and tool use; they must also support explicit orchestration, spatial validation, governance, provenance, failure handling, and human decision authority. Prior work introduced ANASSA, an agentic AI orchestration framework for spatia...
Constantinos Papantoniou, Brian N. Hilton· 0 citations
Large language models are increasingly used as scientific assistants, yet we still lack reliable ways to evaluate whether they can investigate a system rather than merely answer a prompt. A useful evaluation must let agents choose experiments, interpret finite evidence, and make causal decisions, while preserving objec...
Yi-Wei Chen, A. Weerasinghe, S. Biswal et al.· 0 citations
An evidence-grounded benchmark is introduced spanning veterinary, pediatric, and older-adult care, and whether AI advisors maintain appropriate care when users express clinically inferior preferences is tested, and the results expose a broader limitation of user-centric alignment in settings with multiple stakeholders.
Priyadarsi Mishra, A. Beniwal, Zi-Yu Yi et al.· 0 citations
Meta-requirements can help organizations implement AI robustness auditing frameworks, assess existing frameworks for completeness, and refine them and are derived from literature and validated and refine it with practitioners.
J. Owotogbe, R. Nasiri, I. Kumara et al.· 0 citations
In Model Context Protocol (MCP) systems, LLMs select tools from metadata rather than from their implementations. Names and descriptions summarize what a tool does, while the parameter schema specifies the fields a tool accepts and how they are organized. Our analysis of metadata for 367,174 MCP tools reveals substantia...
We propose ConicFlow, which frames constrained sampling from a flow matching model at inference time as an optimal control problem that balances deviation from the pretrained trajectory against reward, subject to hard terminal constraints. We discretize this problem and adapt receding-horizon control, optimizing only a...
Shervin Khalafi, Kasra Fallah, Sung Cho et al.· 0 citations
Conditional diffusion models can generate high-quality images that align closely with a given prompt. However, samples generated from different random seeds may remain visually similar, limiting creative exploration. Existing approaches can increase variation, but often require additional optimization or sampling overh...
Tensor parallelism ends every layer with an all-reduce that sits between two dependent matrix multiplications, so its cost cannot be hidden and grows as compute outpaces interconnect bandwidth. Sending 4-bit instead of 16-bit values would cut this cost by nearly 4x, but the all-reduce must add the values it carries, an...
Xingyu Xie, Xiaoyong Xie, Zhouchen Lin et al.· 0 citations
