A benchmark for predicting final output and checkpoint state from source and input alone, extends CRUXEval-style output prediction with paired shorter- and longer-trace inputs and checkpoints inside and after a loop and exposes errors hidden by short-output scores alone.
Xiao-Hong Chen, David Bucur, Cheng-Long Ma et al.· 0 citations
Proof auto-formalization translates natural-language (NL) theorems and proofs into a formal language (FL) such as Lean, enabling mechanical verification. Despite rapid progress, research-level proofs often depend on concepts missing from leading proof assistant libraries (e.g., Lean's Mathlib), and successful compilati...
Prithwish Jana, Việt Bách Hoàng, Logan Luna et al.· 0 citations
Coding agents need to establish that a program satisfies a specification and that the specification captures the requested behavior. We study how expert diagnosis of verification failures can become reusable guidance for this work. Our approach combines executable language definitions in the K framework with a kit of p...
ModalFidelity, a lightweight router that previews each window and decides, before any forensic detector runs, which stream is worth reading, under a hard compute budget it can never exceed, is presented.
Oğuzhan Başer, Kaan Kale, Sriram Vishwanath et al.· 0 citations
Across twelve models, dense and hybrid, the authors move a hidden state from one story into another where the two routes lead to different places, and read off which place the model gives, which makes the answer a handoff.
Vision-Language-Action (VLA) models have shown strong performance on robotic manipulation, but they often struggle to generalize to unseen tasks, configurations, and long-horizon settings. A key challenge is that VLAs overfit to training scenes and fail to follow novel language instructions. Off-the-shelf vision-langua...