| 2026-11-19 | 2611.24102 | IDEA: Intent-Dominance Evidence Audit for Inference-Time Backdoor Detection in GUI Agents | Zhizhuo Kou et al. | arXiv | 4 | |
| 2026-10-23 | 2610.30882 | Inherited, Not Invented: Tracing the Provenance of Wait in Reasoning Models | Zhizhuo Kou et al. | arXiv | 16 | |
| 2026-10-22 | 2610.21204 | Entertain, then Cancel: How Thinking Makes Reasoning Models Read Some and Or Literally | Zhizhuo Kou et al. | arXiv | 14 | |
| 2026-10-07 | 2610.05166 | A Safe Action Is Not Enough: Feasible-Future Decoding for Vision-Language-Action Policies | Tu Nguyen et al. | arXiv | 3 | |
| 2026-10-05 | 2609.38096 | Tail-Influence Sampling for CVaR Policy Evaluation | Pauline Bourigault et al. | arXiv | 7 | |
| 2026-06-09 | 2606.06735 | A Geometric Account of Activation Steering through Angle-Norm Decomposition | Georgii Aparin et al. | arXiv | 26 | |
| 2026-06-09 | 2606.07473 | Whisper Hallucination Detection and Mitigation via Hidden Representation Steering and Sparse AutoEncoders | Georgii Aparin et al. | arXiv | 14 | |
| 2026-06-04 | 2606.01804 | SpeechEditBench: A Bilingual Multi-Attribute Benchmark for Instruction-Guided Speech Editing | Hanlin Zhang et al. | arXiv | 2 | |