Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
Drafting Agents, Not Just Tokens: Speculative Recursive Self-Improvement for Agentic Systems
Xin Xu, Ke Cheng, Hao Wang +11
Improving an agent through successive revisions requires balancing exploration and evaluation under a limited budget. A narrow search can miss useful sequences of edits, while full…
cs.LG2026
PRISM: Parallel Residual Iterative Sequence Model
Jie Jiang, Ke Cheng, Xin Xu +8
Generative sequence modeling faces a fundamental tension between the expressivity of Transformers and the efficiency of linear sequence models. Existing efficient architectures are…
cs.LG2026
Breaking the Curse of Repulsion: Remoteness-Aware Control of Negative Off-Policy Updates
Yusen Huo, Changping Wang, Yangru Huang +2
Off-policy policy optimization reuses historical behavior, including negative-advantage samples that suppress known failures. We show that repeated reuse can turn this useful signa…