Kelvin Luu

activity

20202026

collaborators

Showing 2026Show all

4 papers · 1 filter

cs.AI2026

BARC-Bench: Benign Asks, Risky Consequences

Kangkang Li, Kelvin Luu, Lily Ma +3

Task instructions from users for agents are inherently underspecified, especially regarding safety boundaries--creating behavioral risks. To study these risks, we introduce BARC-Be…

cs.AI2026

Maintaining Benchmarks Against Increasingly Capable Agents: Detection and Remediation of Unearned Passes

Weijun Luo, Kelvin Luu, Xinyi Liu +6

Agentic benchmarks guide model selection and training. Yet an agent can pass a task without demonstrating the intended capability. Such outcomes constitute unearned passes; their p…

cs.CG2026

Geometric $(1 + ε)$ -Spanners with Few Crossings

Kelvin Luu, Csaba D. Tóth

For $n$ points in the plane and $ε > 0$ , we construct a $(1 + ε)$ -spanner with $O (n / ε)$ edges in which every edge has $O ~ (1/ ε^{3})$ crossin…

cs.AI2026

HiL-Bench (Human-in-Loop Benchmark): Do Agents Know When to Ask for Help?

Tu Trinh, Mohamed Elfeki, Guangze Luo +9

Frontier coding agents solve complex tasks when given complete context but collapse when specifications are incomplete or ambiguous. The bottleneck is not raw capability, but judgm…

BARC-Bench: Benign Asks, Risky Consequences

Maintaining Benchmarks Against Increasingly Capable Agents: Detection and Remediation of Unearned Passes

Geometric (1+ε)-Spanners with Few Crossings

HiL-Bench (Human-in-Loop Benchmark): Do Agents Know When to Ask for Help?

Geometric $(1 + ε)$ -Spanners with Few Crossings