JY
Jingqi Ye
cs.CLcs.AIcs.LGcs.CV
On Valency
published · living versionsW_ap69b7ug·v1 · currentpublished
Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition
with Wenjie Lou, Tianshuo Peng, Shiyang Feng, Shengji Tang +94
1 version
W_fchznfbw·v1published
Preprints & journals
10 papers in the corpus · 2025–2026Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition2610.28578v1 · Wenjie Lou, Tianshuo Peng, Shiyang Feng et al.2026 · 0 citationsarXivon Valency
An Empirical Study of Parameter-Efficient Fine-Tuning for Mixture-of-Experts LLMs2610.26370v1 · Minglei Li, Peng Ye, Haonan He et al.2026 · 0 citationsarXivon Valency
Rethinking Token Reweighting for SFT: Suppress, Reverse, and Extrapolate Learned Features2609.33463v1 · Cunchun Li, Haonan He, Yifan Gao et al.2026 · 0 citationsarXiv
MoE-LoRA: When MoE Models Meet MoE-style Low-Rank Adaptation2607.21978v1 · Qingyu Yang, Haonan He, Minglei Li et al.2026 · 0 citationsarXiv
Parametric Skills2606.30015v1 · Xuan Zhao, Haonan He, Qingyu Yang et al.2026 · 0 citationsarXiv
Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent2606.30616v2 · Lei Bai, Zongsheng Cao, Yang Chen et al.2026 · 0 citationsarXiv
Can Multimodal Large Language Models Truly Understand Small Objects?2604.22884v1 · Fujun Han, Junan Chen, Xintong Zhu et al.2026 · 0 citationsarXiv
A Unified Study of LoRA Variants: Taxonomy, Review, Codebase, and Empirical Evaluation2601.22708v1 · Haonan He, Jingqi Ye, Minglei Li et al.2026 · 0 citationsarXiv
Beyond Gemini-3-Pro: Revisiting LLM Routing and Aggregation at Scale2601.01330v2 · Shengji Tang, Weihao Lin, Peng Ye et al.2026 · 0 citationsarXiv
SCI-Verifier: Scientific Verifier with Thinking2509.24285v1 · Shenghe Zheng, Chenyu Huang, Fangchen Yu et al.2025 · 0 citationsarXiv
Career total: 13 works. 10 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.