InterSketch: An Interleaved Reasoning Model with Self-correcting Visual Sketch and Stepwise Reward 事件

InterSketch: An Interleaved Reasoning Model with Self-correcting Visual Sketch and Stepwise Reward arXiv:2605.26520v1 Announce Type: new Abstract: While vision-language models (VLMs) have exhibited multi-turn visual reasoning capabilities, their reasoning trajectories remain relatively shallow and are dominated by a text-centric paradigm, limiting their applicability to complex visual challenges. In contrast, human-like thought typically involves long-horizon reasoning with an interleaved visua