MG-RWKV: Multi-Grained Context-Aware RWKV for Temporal Forgery Localization
2026-07-01 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the challenge of finding exactly where in audio-visual content manipulations happen, which is hard because current methods are either slow or miss details. They created MG-RWKV, a method that efficiently processes entire sequences by using a new bidirectional model structure without heavy computation. Their approach adapts to different lengths of manipulated segments and reduces mistakes in detecting real parts. Tests on multiple datasets show that their method is both accurate and faster than existing ones.
Temporal Forgery LocalizationArtificial Intelligence-Generated ContentRWKV modelBidirectional architectureMixture of ExpertsCross-Granularity ConsistencyComputational complexityFeature pyramidAudio-visual manipulation
Authors
Jingchen Ni, Cangjin Yu, Dan Jiang, Quan Zhang, Keyu Lv, Shannan Yan, Linyue Pan, Ke Zhang, Chun Yuan
Abstract
Driven by Artificial Intelligence-Generated Content (AIGC), the authenticity of audio-visual content is facing severe challenges. Temporal Forgery Localization (TFL) aims to precisely identify manipulated segments within untrimmed sequences. However, existing methods are limited by CNNs' local receptive fields or Transformers' quadratic complexity, while emerging linear models often struggle to balance global authentic context compression with local abrupt forgery perception. To address this, we propose MG-RWKV, a multi-granularity framework that leverages the data-dependent state evolution of RWKV to achieve efficient full-sequence processing with O(T) complexity. Our framework features three core innovations: (1) a Bidirectional RWKV architecture that captures bidirectional temporal contexts without quadratic overhead; (2) a Multi-Granularity Mixture of Experts (MG-MoE) that performs dynamic routing over explicit temporal receptive fields, adaptively selecting granularities based on forgery duration to significantly enhance decision interpretability; and (3) Cross-Granularity Consistency (CGC), which aligns adjacent feature pyramid levels through hierarchical scale-wise pairing and spatial boundary-aware weighting, effectively reducing false positives in authentic regions. Extensive experiments on Lav-DF, TVIL, and Psynd datasets demonstrate that MG-RWKV achieves state-of-the-art performance with low computational cost.