1 citations · 1 across the 28 of their papers we have counts for
30 papers
HuatuoGPT-3: RL-Only Domain Adaptation from Base Models
Junying Chen, Xinyuan Xie, Ziniu Li +7
Domain adaptation aims to turn a general-purpose large language model (LLM) into an expert for a target domain. While the dominant SFT+RL pipeline offers a convenient cold start, i…
When Do Larger Batches Help Scale LLM Reinforcement Learning?
Ziniu Li, Jinbo Wang, Guanhua Huang +3
Larger batches reduce the variance of stochastic gradients per update and are therefore often expected to accelerate training. Yet whether this statistical benefit translates into…
GameCraft-Bench: Can Agents Build Playable Games End-to-End in a Real Game Engine?
Tongxu Luo, Rongsheng Wang, Jiaxi Bi +22
Game generation is an emerging application of coding agents, requiring models to transform natural-language specifications into playable interactive systems. Unlike traditional cod…
Knowledge Index of Noah's Ark
Sheng Jin, Minghao Liu, Yunze Xiao +24
Knowledge benchmarks for LLMs face three issues: scaling-driven designs that do not operationalize disciplinary representativeness; flat-payment annotation that permits lazy consen…
PrefixFlow: Training-Time KV Caching for Schedule-Level Prefix Reuse in LLM RL Training
Pengbo Li, Feiyuan Zhang, Guangming Sheng +8
In reinforcement learning (RL) post-training, one prompt fans out into trajectories, yet standard trainers replay the shared prefix's forward and backward pass times; in lo…
Balanced Aggregation: Understanding and Fixing Aggregation Bias in GRPO
Zhiyuan Zeng, Jiameng Huang, Zhangyue Yin +8
Reinforcement learning with verifiable rewards (RLVR) has become a central paradigm for improving reasoning and code generation in large language models, and GRPO-style training is…
Do Phone-Use Agents Respect Your Privacy?
Zhengyang Tang, Ke Ji, Xidong Wang +19
We study whether phone-use agents respect privacy while completing benign mobile tasks. This question has remained hard to answer because privacy-compliant behavior is not operatio…
Off-Policy Value-Based Reinforcement Learning for Large Language Models
Peng-Yuan Wang, Ziniu Li, Tian Xu +8
Improving data utilization efficiency is critical for scaling reinforcement learning (RL) for long-horizon tasks where generating trajectories is expensive. However, the dominant R…
Beyond Precision: Training-Inference Mismatch is an Optimization Problem and Simple LR Scheduling Fixes It
Yaxiang Zhang, Yingru Li, Jiacai Liu +4
Reinforcement Learning (RL) for training Large Language Models is notoriously unstable. While recent studies attribute this to "training inference mismatch stemming" from inconsist…
The Optimal Token Baseline: Variance Reduction for Long-Horizon LLM-RL
Yingru Li, Jiawei Xu, Ziniu Li +10
Reinforcement Learning (RL) for Large Language Models (LLMs) often suffers from training collapse in long-horizon tasks due to exploding gradient variance. To mitigate this, a base…
Exploration vs Exploitation: Rethinking RLVR through Clipping, Entropy, and Spurious Reward
Peter Chen, Xiaopeng Li, Ziniu Li +3
This paper examines the exploration-exploitation trade-off in reinforcement learning with verifiable rewards (RLVR), a framework for improving the reasoning of Large Language Model…
The Molecular Structure of Thought: Mapping the Topology of Long Chain-of-Thought Reasoning
Qiguang Chen, Yantao Du, Ziniu Li +10
Large language models (LLMs) often fail to learn effective long chain-of-thought (Long CoT) reasoning from human or non-Long-CoT LLMs imitation. To understand this, we propose that…
Encyclo-K: Evaluating LLMs with Dynamically Composed Knowledge Statements
Yiming Liang, Yizhi Li, Yantao Du +14
Benchmarks play a crucial role in tracking the rapid advancement of large language models (LLMs) and identifying their capability boundaries. However, existing benchmarks predomina…
A Note on Hybrid Online Reinforcement and Imitation Learning for LLMs: Formulations and Algorithms
Yingru Li, Ziniu Li, Jiacai Liu
We present a unified framework for Large Language Model (LLM) fine-tuning that integrates Imitation Learning and Reinforcement Learning. By analyzing the gradient of a composite ob…
Beyond Token-level Supervision: Unlocking the Potential of Decoding-based Regression via Reinforcement Learning
Ming Chen, Sheng Tang, Rong-Xi Tan +4
Decoding-based regression, which reformulates regression as a sequence generation task, has emerged as a promising paradigm of applying large language models for numerical predicti…
Dynamic Vocabulary Pruning: Stable LLM-RL by Taming the Tail
Yingru Li, Jiawei Xu, Jiacai Liu +6
Reinforcement Learning (RL) for Large Language Models (LLMs) faces a fundamental tension: the numerical divergence between high-throughput inference engines and numerically precise…
Trust Region Masking for Long-Horizon LLM Reinforcement Learning
Yingru Li, Jiacai Liu, Jiawei Xu +4
Policy gradient methods for Large Language Models optimize a policy via a surrogate objective computed from samples of a rollout policy . However, modern LLM…
SpeechJudge: Towards Human-Level Judgment for Speech Naturalness
Xueyao Zhang, Chaoren Wang, Huan Liao +8
Aligning large generative models with human feedback is a critical challenge. In speech synthesis, this is particularly pronounced due to the lack of a large-scale human preference…
ORGEval: Graph-Theoretic Evaluation of LLMs in Optimization Modeling
Zhuohan Wang, Ziwei Zhu, Ziniu Li +8
Formulating optimization problems for industrial applications demands significant manual effort and domain expertise. While Large Language Models (LLMs) show promise in automating…
Teaching Language Models to Reason with Tools
Chengpeng Li, Zhengyang Tang, Ziniu Li +8
Large reasoning models (LRMs) like OpenAI-o1 have shown impressive capabilities in natural language reasoning. However, these models frequently demonstrate inefficiencies or inaccu…
Scaling Latent Reasoning via Looped Language Models
Rui-Jie Zhu, Zixuan Wang, Kai Hua +30
Modern LLMs are trained to "think" primarily via explicit text generation, such as chain-of-thought (CoT), which defers reasoning to post-training and under-leverages pre-training…
Knapsack RL: Unlocking Exploration of LLMs via Optimizing Budget Allocation
Ziniu Li, Congliang Chen, Tianyun Yang +5
Large Language Models (LLMs) can self-improve through reinforcement learning, where they generate trajectories to explore and discover better solutions. However, this exploration p…
TreePO: Bridging the Gap of Policy Optimization and Efficacy and Inference Efficiency with Heuristic Tree-based Modeling
Yizhi Li, Qingshui Gu, Zhoufutu Wen +14
Recent advancements in aligning large language models via reinforcement learning have achieved remarkable gains in solving complex reasoning problems, but at the cost of expensive…
Bridging Formal Language with Chain-of-Thought Reasoning to Geometry Problem Solving
Tianyun Yang, Yunwen Li, Ziniu Li +3
Large vision language models exhibit notable limitations on Geometry Problem Solving (GPS) because of their unreliable diagram interpretation and pure natural-language reasoning. A…