Few-step text-to-image generators produce high-quality images with few network forward passes, yet struggle with complex prompts involving multiple subjects, interactions, spatial relations, counting, and attribute binding. Can these errors be reduced without changing the generator's parameters or increasing its forwar...
Ya-Ting Gao, Zheng Zhu, Zi-Hao Wen et al.· 0 citations
Outcome-Guided On-Policy Distillation is proposed, which applies trajectory-relative weighting to teacher supervision and calibrates these weights using final task outcomes from paired student continuations and selectively strengthens supervision on the student's original trajectories at turns where teacher guidance be...
Tong Zhang, Zhou Liu, Yi-Hao Liu et al.· 0 citations
This work presents Token-Disentangled Latent Test-Time Scaling, an inference-time framework that makes latent refinement token-role-aware and outperforms strong output-space test-time scaling baselines under matched decoded-candidate budgets.
Hao-Xuan Ma, Yi-Hao Liu, Yu-Tao Sun et al.· 0 citations
ConSPO is a Contrastive Sequence-level Policy Optimization method that uses length-normalized sequence log-probabilities as rollout scores and contrasts verified positive rollouts against negative distractors within the same group, and outperforms strong baselines on challenging reasoning benchmarks.
Generative query suggestion aims to enhance user engagement by anticipating user intents and recommending relevant follow-up queries. A central challenge is to generate slates whose individual queries are useful while the slate covers distinct intents. We propose an Intent-Driven Query Suggestion Framework with dual-st...
Xin-Peng Liu, Lu-Yao Ma, Jia-Yi Qiao et al.· 0 citations
Elastic Horizon is introduced, a closed-loop controller that tracks a dynamic boundary beyond which additional interactions yield diminishing returns while cost grows linearly and stabilizes the horizon inside the saturation band from both under- and over-capacity initializations.
Gang-Yi Zhang, Jun-Jie Meng, Le-Tian Zhang et al.· 0 citations
Specialize-and-Merge Online Policy Distillation (SMOPD) is proposed, a two-stage training method for multi-reward optimization that outperforms GDPO across 1.5B, 3B and 7B backbones.
Wen Wang, Jia-Hua Bao, Tu Yongsiqi et al.· 1 citation
NapMem is introduced, a framework for learning to use long-term user memory as a structured action space rather than passively retrieved context, and suggests that long-term user memory benefits from coupling structured storage with a learned policy for using memory at the appropriate granularity.
This work introduces Skill Self-Play (Skill-SP), a co-evolutionary framework comprising a proposer, a solver, and a dynamic skill controller that effectively bridges the gap between structured verification and open-ended exploration.
Siyuan Huang, Pengyu Cheng, Haotian Liu et al.· arXiv.org· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
