Rewrites text-to-image prompts with reinforcement learning that aligns policy updates to semantic spans and pools log-probability shifts within each span to weight token losses. Improves prompt engineering for frozen generators without extra renders (≈9.5% on GenEval, 7.6% on T2I-CompBench).
AI Agent Papers2026
Turns a VLM into a robot-coding agent that directly drives a dual-arm robot from three RGB cameras, iteratively writing and fixing small executable perception and control tools from observations and logs; evaluated on 42 RoboDojo tasks.
AI Agent Papers2026
Alternates harness and policy optimization so a coding agent uses its own policy to propose harness revisions that target evolving failure modes. Combines harness search (H-step) with RL-based policy updates (P-step), improving resolve rates on multiple coding benchmarks and generalizing out-of-distribution.
Improves stability and consistency for multi-turn image editing by training autoregressive diffusion editors with Single-Turn Distribution Matching Distillation and Teacher-Forcing Guidance. Combines a single-turn teacher for visual-quality supervision and trajectory-based teacher-forcing to preserve edits across successive turns.