| Inconsistency-aware Multimodal Schrodinger Bridge for Deepfake Localization | | 99.33 | 95.52 | 55.92 | 94.68 | 95.41 | 95.5 | 95.62 | IaMSB | 2026-05-22 |
| MG-RWKV: Multi-Grained Context-Aware RWKV for Temporal Forgery Localization | | 98.92 | 93.41 | 38.47 | | | | | MG-RWKV (multimodal; full set) | 2026-07-01 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 96.3 | 81.62 | 4.44 | 78.75 | 79.4 | 80.48 | 84.96 | BA-TFD+ (audio-visual) | 2023-05-03 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 86.33 | 74.47 | 3.05 | 74.45 | 74.46 | 74.47 | 70.23 | TriDet | 2023-05-03 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 85.23 | 77.23 | 0.93 | 76.93 | 77.19 | 77.23 | 59.05 | ActionFormer | 2023-05-03 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 80.22 | 72.5 | 5.22 | 69.18 | 70.56 | 72.5 | 61.04 | TadTR | 2023-05-03 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 79.15 | 67.03 | 0.24 | 58.51 | 60.89 | 64.18 | 38.57 | BA-TFD | 2023-05-03 |
| Do You Really Mean That? Content Driven Audio-Visual Deepfake Dataset and Multimodal Method for Temporal Forgery Localization | ✓ Link | 76.90 | 66.90 | 0.25 | 58.42 | 60.77 | 64.08 | 38.50 | BA-TFD (multimodal; full set) | 2022-04-13 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 65.38 | 62.98 | 0.11 | 52.11 | 54.8 | 59.26 | 23.89 | AVFusion | 2023-05-03 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 64.78 | 64.0 | 2.53 | 54.38 | 55.94 | 59.33 | 54.85 | BA-TFD+ (visual-only) | 2023-05-03 |
| Do You Really Mean That? Content Driven Audio-Visual Deepfake Dataset and Multimodal Method for Temporal Forgery Localization | ✓ Link | 58.55 | 62.49 | 0.16 | 50.29 | 53.86 | 58.77 | 28.60 | BA-TFD (visual-only; full set) | 2022-04-13 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 56.41 | 74.93 | 0.21 | 59.29 | 64.98 | 71.11 | 32.57 | BSN++ | 2023-05-03 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 24.01 | 53.26 | 0.07 | 26.93 | 31.6 | 41.24 | 7.61 | BMN (E2E) | 2023-05-03 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 17.85 | 43.15 | 0.11 | 16.71 | 24.59 | 34.23 | 9.42 | AGT | 2023-05-03 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 12.78 | 37.88 | 0.0 | 32.15 | 34.39 | 36.71 | 1.62 | MDS | 2023-05-03 |
| Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization | ✓ Link | 10.56 | 48.49 | 0.0 | 31.55 | 37.13 | 44.39 | 1.66 | BMN | 2023-05-03 |