Searcharxiv⌕ Search

arXiv subjects

Xinmiao Luan

Publications and source records attributed to Xinmiao Luan.

3 recordsLinked to original sources

Reality Perception Gap: Are AI-Manipulated Video Detectors Ready for the Real World?

AI-manipulated video detectors are typically evaluated on clean, curated datasets. In deployment, however, videos have usually passed through repost compression on social-media platforms and sometimes were captured by imperfect hardware. To measure detection under these conditions, we introduce Reality Perception Gap (RPG), a benchmark of 10,000 videos from frontier generators and partially synthetic edits, processed with cross-platform repost compression and paired with real footage that carries camera-side optical artifacts. RPG also includes socially fragile content, such as crime and wartime footage, where fabricated videos can cause severe social harm. Across 18 detectors, none catches more than 15% of fakes at a 1% false-positive rate. A positive control and a re-encoding sweep separate two sources of failure: generator and content shift degrades detectors before any compression, and repost compression erases the low-level traces that the artifact-based detectors relies on. Existing detectors therefore struggle when deployment conditions strip away the generation artifacts they rely on. These complementary failures suggest that robust detection requires training under deployment conditions and combining low-level, temporal, and semantic evidence, where passive detection may have reached its limits.

cs.CV↗

My FAULT: Self-Diagnosis as Credit Assignment in Self-Evolving Agentic Reinforcement Learning

Agentic reinforcement learning (RL) has emerged as a powerful approach for training large language model agents on multi-step tasks, yet reliance on terminal outcome rewards creates two credit-assignment problems, particularly in long-horizon tasks. First, same-outcome rollout groups provide no learning signal from terminal rewards. Second, terminal rewards provide only trajectory-wide feedback, making it difficult to identify which decisions caused a failure. Recent work supplements terminal rewards with finer-grained information from trajectory analysis, such as natural-language reflections on intermediate decisions and errors. However, natural-language diagnoses are difficult to use directly for credit assignment: their error claims may be unreliable, and they do not quantify how much each error should affect learning. We propose Self-Diagnosis-guided Terminal Credit Redistribution (FAULT), which turns diagnosed errors into explicit step-level credit anchored by terminal outcomes. FAULT checks diagnostic evidence and learns relative error costs from task outcomes. During training, the policy and self-diagnoser co-evolve, while error costs are updated online from recent outcomes. On ALFWorld, FAULT recovers learning signals from same-outcome groups, reaching 95% signal coverage versus 41% for GRPO and 72% for GiGPO, while better localizing credit to specific error steps. Across two model scales, FAULT delivers strong. improvements on the long-horizon ALFWorld and WebShop tasks while remaining competitive on short-horizon Search-based QA.

cs.CL↗

Efficient online cross-covariance monitoring with incremental SVD: An approach for the detection of emerging dependency patterns in IoT systems

The development of the manufacturing systems has made it increasingly necessary to monitor the data generated by multiple interconnected subsystems with rapid incoming of samples. Based on incremental Singular Value Decomposition (ISVD), we develop a general online monitoring approach for the relationship of data generated from two interconnected subsystems, where each subsystem produces big data streams with several variation patterns in normal working condition. When special situation happens and new associations occur, a very small amount of computation is sufficient to update the system status and compute the control statistics by using this approach. The proposed method reduces computational overhead and retains only a small number of pairs of possible dependent patterns at each step. The validation of the method through simulation studies and a case study on semiconductor manufacturing processes further supports its effectiveness.

eess.SY↗