1 citations · 3 across the 15 of their papers we have counts for
12 papers · 1 filter
From Refuse to Richness: Rubric Rewards for Long-Form Hallucination Reinforcement Learning
Yudong Wang, Zhe Yang, Wenhan Ma +6
Rewards that penalize unsupported claims can improve grounding in long-form generation, but they can also teach models to answer less. We study this refusal-to-richness trade-off i…
MOPD: Multi-Teacher On-Policy Distillation for Capability Integration in LLM Post-Training
Wenhan Ma, Jianyu Wei, Liang Zhao +10
Modern large language models (LLMs) rely on reinforcement learning during post-training to push specific capabilities, yet integrating multiple capabilities into one model remains…
Sparse-BitNet: 1.58-bit LLMs are Naturally Friendly to Semi-Structured Sparsity
Di Zhang, Xun Wu, Shaohan Huang +9
Semi-structured N:M sparsity and low-bit quantization (e.g., 1.58-bit BitNet) are two promising approaches for improving the efficiency of large language models (LLMs), yet they ha…
Towards Better RL Training Data Utilization via Second-Order Rollout
Zhe Yang, Yudong Wang, Rang Li +1
Reinforcement Learning (RL) has empowered Large Language Models (LLMs) with strong reasoning capabilities, but vanilla RL mainly focuses on generation capability improvement by tra…
Enhancing Reliability across Short and Long-Form QA via Reinforcement Learning
Yudong Wang, Zhe Yang, Wenhan Ma +2
While reinforcement learning has unlocked unprecedented complex reasoning in large language models, it has also amplified their propensity for hallucination, creating a critical tr…
Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers
Wenhan Ma, Hailin Zhang, Liang Zhao +4
Reinforcement learning (RL) has emerged as a crucial approach for enhancing the capabilities of large language models. However, in Mixture-of-Experts (MoE) models, the routing mech…
A Probabilistic Inference Scaling Theory for LLM Self-Correction
Zhe Yang, Yichang Zhang, Yudong Wang +3
Large Language Models (LLMs) have demonstrated the capability to refine their generated answers through self-correction, enabling continuous performance improvement over multiple r…
MiMo: Unlocking the Reasoning Potential of Language Model -- From Pretraining to Posttraining
LLM-Core Xiaomi, :, Bingquan Xia +62
We present MiMo-7B, a large language model born for reasoning tasks, with optimization across both pre-training and post-training stages. During pre-training, we enhance the data p…
MiMo-VL Technical Report
Core Team, Zihao Yue, Zhenru Lin +71
We open-source MiMo-VL-7B-SFT and MiMo-VL-7B-RL, two powerful vision-language models delivering state-of-the-art performance in both general visual understanding and multimodal rea…
From Mathematical Reasoning to Code: Generalization of Process Reward Models in Test-Time Scaling
Zhengyu Chen, Yudong Wang, Teng Xiao +5
Recent advancements in improving the reasoning capabilities of Large Language Models have underscored the efficacy of Process Reward Models (PRMs) in addressing intermediate errors…
Confidence v.s. Critique: A Decomposition of Self-Correction Capability for LLMs
Zhe Yang, Yichang Zhang, Yudong Wang +3
Large Language Models (LLMs) can correct their self-generated responses, but a decline in accuracy after self-correction is also witnessed. To have a deeper understanding of self-c…
A Challenging Benchmark for Low-Resource Learning
Yudong Wang, Chang Ma, Qingxiu Dong +2
With promising yet saturated results in high-resource settings, low-resource datasets have gradually become popular benchmarks for evaluating the learning ability of advanced neura…