Searcharxiv⌕ Search

arXiv subjects

Siyu Gan

Publications and source records attributed to Siyu Gan.

3 recordsLinked to original sources

OverServe: Resource Amplification Attacks on SaaS Agents via Tool--Prompt Co-Design

Advances in large language models (LLMs) are enabling software-as-a-service (SaaS) providers to automate longer business workflows with agents. In subscription-based SaaS deployments, model and tool calls consume provider resources without additional customer payments. Because the agent autonomously decides when to stop, a malicious tenant can send a prompt to the agent, exploiting its completion decisions to extend execution and waste provider resources. In this paper, we present OverServe, a multi-agent framework for constructing such attacks against SaaS agents. OverServe co-designs Model Context Protocol (MCP) status tools and completion-confirmation prompts for the attacker. Each tool call returns accurate status and terminates normally, yet the agent may repeatedly check an unfinished condition to satisfy the prompt. This consumes the SaaS provider's resources even after the original business task is completed. OverServe constructs service-specific tool--prompt pairs and uses separate tool and prompt reviews to guide candidate pair selection. Tested tool specifications trigger no alerts in the evaluated MCP scanners. We evaluate Qwen, Nemotron, DeepSeek, and Sonnet on selected tasks from nine benchmarks. All OverServe runs pass the original task checks. OverServe can reach a maximal amplification of API requests by $93.88\times$ while preserving the original task goal and collateral checks.

cs.CR↗

BridgeCast: Bridging Ocean Wave Forecasts to Reanalysis via Flow Matching with Exogenous Variables

Ocean wave forecasting is essential for maritime safety, offshore operations, and coastal resilience, yet remains challenging due to systematic biases in physics-based models. Physical models, while widely used, rely on approximations and parameterizations that limit their accuracy under complex ocean-atmosphere conditions. To enhance ocean wave forecasting, we propose BridgeCast, within a physics-AI hybrid framework for bias correction. BridgeCast is a probabilistic model based on conditional flow matching (CFM) that learns to transform physical model forecasts into reanalysis-like fields. It treats physical forecasts as corrupted observations and employs a continuous-time generative process to bridge their distribution toward that of reanalysis data. BridgeCast is parameterized by a Transformer-based architecture that enables spatiotemporal modeling, incorporation of exogenous atmospheric variables, and flexible inference via both ordinary and stochastic differential equation formulations. Extensive experiments on real-world datasets demonstrate that BridgeCast consistently outperforms state-of-the-art baselines across regions and forecast lead times.

physics.ao-ph↗

Chip games and multipartite graph paintability

We study the paintability, an on-line version of choosability, of complete multipartite graphs. We do this by considering an equivalent chip game introduced by Duraj, Gutowski, and Kozik. We consider complete multipartite graphs with $ n $ parts of size at most 3. Using a computational approach, we establish upper bounds on the paintability of such graphs for small values of $ n. $ The choosability of complete multipartite graphs is closely related to value $ p(n, m) $, the minimum number of edges in a $n$-uniform hypergraph with no panchromatic $m$-coloring. We consider an online variant of this parameter $ p_{OL}(n, m), $ introduced by Khuzieva et al. using a symmetric chip game. With this symmetric chip game, we find an improved upper bound for $ p_{OL}(n, m)$ when $m \geq 3$ and $n$ is large. Our method also implies a lower bound on the paintability of complete multipartite graphs with $m \geq 3$ parts of equal size.

math.CO↗