Publications (23)
TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning
Heming Zou, Qi Wang, Yun Qu +9
Reinforcement learning with verifiable rewards (RLVR) is a promising approach for enhancing reasoning and agentic behavior in large language models. However, rollout-intensive poli…
CAME: Contrastive Automated Model Evaluation
Ru Peng, Qiuyang Duan, Haobo Wang +5
The Automated Model Evaluation (AutoEval) framework entertains the possibility of evaluating a trained machine learning model without resorting to a labeled testing set. Despite th…
Riemann-Stieltjes operators and multipliers on spaces in the unit ball of
Ru Peng, Caiheng Ouyang
This paper is devoted to characterizing the Riemann-Stieltjes operators and pointwise multipliers acting on Mbius invariant spaces , which unify BMOA and Bloch…
Can LLM design high-quality experiments? A Comprehensive and Systematic Benchmark on Autonomous Experimental Design
Zejun Liu, Jian Wu, Ru Peng +4
AI for Research (AI4Research) leverages AI to automate and improve scientific workflows. While experimental design is a critical stage of the research process, prior work has focus…
Distill the Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation
Ru Peng, Yawen Zeng, Junbo Zhao
Past works on multimodal machine translation (MMT) elevate bilingual setup by incorporating additional aligned vision information. However, an image-must requirement of the multimo…
Baryonic Effect on chi_cJ Suppression in Au+Au Collisions at RHIC Energies
Ru Peng, Xiao-Ming Xu, Dai-Cui Zhou
We predict that initially produced chi_cJ mesons at low transverse momentum in the central rapidity region are almost dissociated by nucleons and antinucleons in hadronic matter pr…