Measuring Robustness and Efficiency in a Connectome-Constrained Fly Visual System Model: Linking an Adversarial and a Perceptual Illusion

Summary

Researchers benchmarked FlyNet, a connectome-constrained recurrent model of the fly visual system, against CNNs, RAFT, and CTRNN variants on a collision-detection task. FlyNet is adversarially vulnerable but more robust to common video corruptions than RAFT relative to its compute, using up to five orders of magnitude fewer FLOPs. The authors link its adversarial failures to the reverse-phi illusion through a shared low-level motion statistic.

Related items

• InfoImage watermarking robustness under MLLM-guided semantic editing: Benchmark, defense, and diagnosisSimilar attack · Elsevier Security Journals
• LowYour Robot's Safety Functions Already Work. What If the Input Lies?Similar attack · VicOne Blog
• InfoSpeech Tokenizers are Vulnerable: Transferable Semantic Attack and Robust TokenizerSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoAdversarial Retraining for Machine Learning-Based Man-in-the-Middle Attack Detection in Smart Grid Industrial Control SystemsSimilar attack · OpenAlex (peer-reviewed AI security)
• LowMeasuring Robustness and Efficiency in a Connectome-Constrained Fly Visual System Model: Linking an Adversarial and a Perceptual IllusionSimilar attack · Arxiv (cs.RO + cs.CV security)