8 papers
Rethinking Token Reweighting for SFT: Suppress, Reverse, and Extrapolate Learned Features
Cunchun Li, Haonan He, Yifan Gao +4
Supervised fine-tuning (SFT) learns most aggressively from tokens that the model deems least likely. This helps acquire new behaviors, but also amplifies noisy or conflicting super…
MoE-LoRA: When MoE Models Meet MoE-style Low-Rank Adaptation
Qingyu Yang, Haonan He, Minglei Li +4
Mixture-of-Experts (MoE) architectures have been widely adopted in large language models, yet parameter-efficient fine-tuning (PEFT) for MoE models remains underexplored. Existing…
Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent
Lei Bai, Zongsheng Cao, Yang Chen +50
We introduce Agents-A1, a 35B Mixture-of-Experts Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. We investigate agent-horizon scaling…
Parametric Skills
Xuan Zhao, Haonan He, Qingyu Yang +5
Since intelligence fundamentally relies on efficient skill acquisition (Chollet, 2019), the ability to leverage skills is critical. For LLMs, skills, manually authored or extracted…
Can Multimodal Large Language Models Truly Understand Small Objects?
Fujun Han, Junan Chen, Xintong Zhu +4
Multimodal Large Language Models (MLLMs) have shown promising potential in diverse understanding tasks, e.g., image and video analysis, math and physics olympiads. However, they re…
Beyond Gemini-3-Pro: Revisiting LLM Routing and Aggregation at Scale
Shengji Tang, Weihao Lin, Peng Ye +9
Large Language Models (LLMs) have rapidly advanced, with Gemini-3-Pro setting a new performance milestone. In this work, we explore collective intelligence as an alternative to mon…
A Unified Study of LoRA Variants: Taxonomy, Review, Codebase, and Empirical Evaluation
Haonan He, Jingqi Ye, Minglei Li +4
Low-Rank Adaptation (LoRA) is a fundamental parameter-efficient fine-tuning method that balances efficiency and performance in large-scale neural networks. However, the proliferati…
SCI-Verifier: Scientific Verifier with Thinking
Shenghe Zheng, Chenyu Huang, Fangchen Yu +8
As large language models (LLMs) are increasingly applied to scientific reasoning, the complexity of answer formats and the diversity of equivalent expressions make answer verificat…