Inference-time search can substantially improve LLM reasoning when tasks admit deterministic verification, but existing methods largely refine single trajectories and lack a reliable mechanism for composing partial solutions across candidates.We propose contractchecked graph editing: represent each candidate as an interface-typed reasoning DAG and validate every nontrivial edit with a deterministic structural gate (acyclicity, namespace closure, schema validity, terminal constraints) before invoking the verifier.The gate certifies runnability only and emits auditable rejection reasons; semantic correctness is determined solely by the verifier.Instantiated in Genetic Inference Search (GIS) with Qwen2.5-32B-Instructunder strictly matched token budgets (8K tokens), contractchecked grafting increases verifier-runnable recombination from 41.2% to 92.8% and improves accuracy over rStar (+6.1 on MATH, +9.1 on MATH L5) while using 42% fewer verifier calls.The same operators transfer across outer loops (beam, best-first, MCTS) and to structured generation and code, outperforming execution-guided beam search on Spider (+2.8) and improving multi-file code generation on HumanEval-MF (+9.2).

The proliferation of Unmanned Aerial Vehicles (UAVs) in 6G-enabled edge computing presents a dual challenge of ensuring secure data transmission and performing accurate object detection on resource-constrained platforms. This paper proposes a Synergistic Encryption and Detection framework (SED-UAV) that integrates a lightweight chaotic image encryption algorithm, LB-ICA, and an enhanced small object detector, YOLO-LFP. The LB-ICA algorithm leverages an improved chaotic map and a plaintext-aware key mechanism using SHA-256. It achieves a key space of over 2100 and an average encryption speed of 20.3 ms per block, providing robust security against differential and chosen-plaintext attacks with high efficiency suitable for UAVs. For detection, the YOLO-LFP model enhances the YOLOv8 baseline with a hybrid attention module, an adaptive feature fusion strategy, and a dedicated small object head. Experimental results on the VisDrone2019 benchmark show YOLO-LFP achieves a state-of-the-art performance of 44.7% [email protected], significantly outperforming existing models. This work provides a comprehensive solution for deploying secure, real-time computer vision applications on UAV platforms.

Data scarcity is a very common real-world problem that poses a major challenge to data-driven analytics. Although a lot of data-balancing approaches have been proposed to mitigate this problem, they may drop some useful information or fall into the overfitting problem. Generative Adversarial Network (GAN) based data synthesis methods can alleviate such a problem but lack of quality control over the generated samples. Moreover, the latent associations between the attribute set and the class labels in a relational data cannot be easily captured by a vanilla GAN. In light of this, we introduce an end-to-end self-training scheme (namely, Quality-Aware Self-Training) for rare relational data synthesis, which generates labeled synthetic data via pseudo labeling on GAN-based synthesis. We design a semantic pseudo labeling module to first control the quality of the generated features/samples, then calibrate their semantic labels via a classifier committee consisting of multiple pre-trained shallow classifiers. The high-confident generated samples with calibrated pseudo labels are then fed into a semantic classification network as augmented samples for self-training. We conduct extensive experiments on 20 benchmark datasets of different domains, including 14 industrial datasets. The results show that our method significantly outperforms state-of-the-art methods, including two recent GAN-based data synthesis schemes. Codes are available at https://github.com/yaxinhou/QAST.

Sensitive resources in Trusted Execution Environment (TEE) have suffered serious security threats in recent years. Previous protection approaches either lack a strong assurance of TEE security properties or are limited to a single platform. We propose a compatible verified TEE architecture, calledCVTEE, which delegates a security monitor to manage TEE resources securely. This architecture has two key advantages: i) its functional correctness and security are guaranteed by a machine-checkable proof of security objectives of Trusted Application (TA) isolation, runtime confidentiality, and runtime integrity, and ii) it is applicable to different TEE platforms and implementation-independent due to its high level of abstraction and non-determinism of data types. Note that access control policy and information flow control policy are the core for security management of resources. After formally specifying the security attributes of TEE resources, we develop these policies based on Common Criteria (CC) in the security monitor and provide atomic interfaces.CVTEEis formally verified with 386 lemmas/theorems and$\sim$10,000 LOC of Isabelle/HOL. In addition, we implement a proof of concept for the access control module of Teaclave, and prove that the constructed access control model meets the security requirements through 5 theorems.

Oracle Bone Inscriptions (OBI) research is very meaningful for both history and literature. In this paper, we introduce our contributions in AI-Powered Oracle Bone (OB) fragments rejoining and OBI recognition. (1) We build a real-world dataset OB-Rejoin, and propose an effective OB rejoining algorithm which yields a top-10 accuracy of 98.39%. (2) We design a practical annotation software to facilitate OBI annotation, and build OracleBone-8000, a large-scale dataset with character-level annotations. We adopt deep learning based scene text detection algorithms for OBI localization, which yield an F-score of 89.7%. We propose a novel deep template matching algorithm for OBI recognition which achieves an overall accuracy of 80.9%. Since we have been cooperating closely with OBI domain experts, our effort above helps advance their research. The resources of this work are available at https://github.com/chongshengzhang/OracleBone.

This paper presents a method of synchronizing video sequences with non-rigid human motion in a surveillance video stitching scenario. Different from manual or half-manual video synchronization methods, our work exploits the spatiotemporal structure information embedded in the time-series data itself to realize synchronization automatically and consistently. After projecting into the same plane based on the homography constraint, we formulate the problem of aligning two time series data with unequal lengths as an improved dynamic time warping (DTW)-based optimization problem. Furthermore, we propose a simple but effective shape registration and similarity measure method based on ℛ transform to decrease shape ambiguities. The method has been evaluated qualitatively on several real sequences and demonstrated its effective performance.