9 papers
Focal Reward: Balanced Reinforcement Learning under Rubric-Based Rewards
Yu Huang, Zihua Zhao, Zhaoxin Huan +9
The open-ended generation in LLMs usually requires multi-dimensional rubrics to adequately assess quality and guide the improvement of reinforcement learning. However, a critical d…
Roll Out and Roll Back: Diffusion LLMs are Their Own Efficiency Teachers
Fanqin Zeng, Feng Hong, Geng Yu +6
Diffusion Large Language Models (DLLMs) promise fast parallel generation, yet open-source DLLMs still face a severe quality-speed trade-off: accelerating decoding by revealing mult…
TriSpec: Ternary Speculative Decoding via Lightweight Proxy Verification
Haoyun Jiang, Junqi He, Feng Hong +8
Inference efficiency in Large Language Models (LLMs) is fundamentally limited by their serial, autoregressive generation, especially as reasoning becomes a key capability and respo…
Innovator: Scientific Continued Pretraining with Fine-grained MoE Upcycling
Ning Liao, Xiaoxing Wang, Zehao Lin +18
A large language model (LLM) with knowledge in both scientific and general tasks is the foundation of science general intelligence. However, directly continued pretraining an LLM u…
Learning to Instruct for Visual Instruction Tuning
Zhihan Zhou, Feng Hong, Jiaan Luo +5
We propose L2T, an advancement of visual instruction tuning (VIT). While VIT equips Multimodal LLMs (MLLMs) with promising multimodal capabilities, the current design choices for V…
Dual-granularity Sinkhorn Distillation for Enhanced Learning from Long-tailed Noisy Data
Feng Hong, Yu Huang, Zihua Zhao +5
Real-world datasets for deep learning frequently suffer from the co-occurring challenges of class imbalance and label noise, hindering model performance. While methods exist for ea…