256 citations · 342 across the 78 of their papers we have counts for
21 papers · 1 filter
WisPaper: Your AI Scholar Search Engine
Li Ju, Jun Zhao, Mingxu Chai +24
We present \textsc{WisPaper}, an end-to-end agent system that transforms how researchers discover, organize, and track academic literature. The system addresses two fundamental cha…
DVPO: Distributional Value Modeling-based Policy Optimization for LLM Post-Training
Dingwei Zhu, Zhiheng Xi, Shihan Dou +15
Reinforcement learning (RL) has shown strong performance in LLM post-training, but real-world deployment often involves noisy or incomplete supervision. In such settings, complex a…
From Scores to Preferences: Redefining MOS Benchmarking for Speech Quality Reward Modeling
Yifei Cao, Changhao Jiang, Jiabao Zhuang +11
Assessing the perceptual quality of synthetic speech is crucial for guiding the development and refinement of speech generation models. However, it has traditionally relied on huma…
Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy Optimization
Boyang Liu, Yifan Hu, Senjie Jin +5
Multimodal large language models (MLLMs) are well suited to image aesthetic assessment, as they can capture high-level aesthetic features leveraging their cross-modal understanding…
CMDAR: A Chinese Multi-scene Dynamic Audio Reasoning Benchmark with Diverse Challenges
Hui Li, Changhao Jiang, Hongyu Wang +11
The ability to reason from audio, including speech, environmental sounds, and music, is essential for AI agents to interact effectively in real-world scenarios. Existing benchmarks…
LLMEval-Fair: A Large-Scale Longitudinal Study on Robust and Fair Evaluation of Large Language Models
Ming Zhang, Yujiong Shen, Jingyi Deng +19
Existing evaluation of Large Language Models (LLMs) on static benchmarks is vulnerable to data contamination and leaderboard overfitting, critical issues that obscure true model ca…
VRPO: Rethinking Value Modeling for Robust RL under Noisy Supervision in LLM Post-Training
Dingwei Zhu, Shihan Dou, Zhiheng Xi +16
Reinforcement Learning (RL) in real-world environments often suffers from ambiguous or incomplete reward supervision, which undermines policy stability and generalization. Such noi…
Alleviating Attention Hacking in Discriminative Reward Modeling through Interaction Distillation
Jianxiang Zang
The reward model (RM), as the core component of reinforcement learning from human feedback (RLHF) for large language models (LLMs), responsible for providing reward signals to gene…
SpeechRole: A Large-Scale Dataset and Benchmark for Evaluating Speech Role-Playing Agents
Changhao Jiang, Jiajun Sun, Yifei Cao +15
Speech is essential for realistic role-playing, yet existing work on role-playing agents largely centers on text, leaving Speech Role-Playing Agents (SRPAs) underexplored and witho…
Pre-Trained Policy Discriminators are General Reward Models
Shihan Dou, Shichun Liu, Yuming Yang +19
We offer a novel perspective on reward modeling by formulating it as a policy discriminator, which quantifies the difference between two policies to generate a reward signal, guidi…
Progressive Mastery: Customized Curriculum Learning with Guided Prompting for Mathematical Reasoning
Muling Wu, Qi Qian, Wenhao Liu +12
Large Language Models (LLMs) have achieved remarkable performance across various reasoning tasks, yet post-training is constrained by inefficient sample utilization and inflexible…
LLMEval-Med: A Real-world Clinical Benchmark for Medical LLMs with Physician Validation
Ming Zhang, Yujiong Shen, Zelin Li +13
Evaluating large language models (LLMs) in medicine is crucial because medical applications require high accuracy with little room for error. Current medical benchmarks have three…
EvaLearn: Quantifying the Learning Capability and Efficiency of LLMs via Sequential Problem Solving
Shihan Dou, Ming Zhang, Chenhao Huang +14
We introduce EvaLearn, a pioneering benchmark designed to evaluate large language models (LLMs) on their learning capability and efficiency in challenging tasks, a critical, yet un…
Two Minds Better Than One: Collaborative Reward Modeling for LLM Alignment
Jiazheng Zhang, Wenqing Jing, Zizhuo Zhang +12
Reward models (RMs) are central to aligning large language models (LLMs) with human values, yet noisy preference feedback can induce reward misgeneralization through spurious corre…
Compression Hacking: A Supplementary Perspective on Informatics Properties of Language Models from Geometric Distortion
Jianxiang Zang, Meiling Ning, Yongda Wei +7
Recently, the concept of ``compression as intelligence'' has provided a novel informatics metric perspective for language models (LMs), emphasizing that highly structured represent…
Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning
ByteDance Seed, :, Jiaze Chen +267
We introduce Seed1.5-Thinking, capable of reasoning through thinking before responding, resulting in improved performance on a wide range of benchmarks. Seed1.5-Thinking achieves 8…
Improving RL Exploration for LLM Reasoning through Retrospective Replay
Shihan Dou, Muling Wu, Jingwen Xu +4
Reinforcement learning (RL) has increasingly become a pivotal technique in the post-training of large language models (LLMs). The effective exploration of the output space is essen…
PFDial: A Structured Dialogue Instruction Fine-tuning Method Based on UML Flowcharts
Ming Zhang, Yuhui Wang, Yujiong Shen +16
Process-driven dialogue systems, which operate under strict predefined process constraints, are essential in customer service and equipment maintenance scenarios. Although Large La…
Detecting Essence Code Clones via Information Theoretic Analysis
Lida Zhao, Shihan Dou, Yutao Hu +8
Code cloning, a widespread practice in software development, involves replicating code fragments to save time but often at the expense of software maintainability and quality. In t…
Measuring Data Diversity for Instruction Tuning: A Systematic Analysis and A Reliable Metric
Yuming Yang, Yang Nan, Junjie Ye +8
Data diversity is crucial for the instruction tuning of large language models. Existing studies have explored various diversity-aware data selection methods to construct high-quali…
Beyond Scaling: Measuring and Predicting the Upper Bound of Knowledge Retention in Language Model Pre-Training
Changhao Jiang, Ming Zhang, Yifei Cao +12
The GPT-4 technical report suggests that downstream performance can be predicted from pre-training signals, but offers little methodological detail on how to quantify this. This wo…