2 citations · 4 across the 16 of their papers we have counts for
20 papers
CleanRM: Reward Modeling from Noisy Feedback with a Few Clean Feedback for RLHF
Licheng Pan, Haocheng Yang, Yuzhou Chen +7
Feedback noise is pervasive in preference data, arising from subjective human judgments, annotation mistakes, and imperfect LLM judges. Standard reward modeling trained on such fee…
Not All Criteria Matter: Importance-Aware Criteria Weighting for Rubric-Based LLM Evaluation
Haocheng Yang, Licheng Pan, Yiming Dong +7
Rubric-based evaluation assesses the responses of large language models (LLMs) by aggregating the verification results of explicit criteria, yet not all criteria matter equally. Ho…
ZYT-World: A Real-Time Controllable World Model for Closed-Loop Autonomous-Driving Simulation
Boni Hu, Xiong Wei, Haoming Huang +20
Generative world models offer controllable and repeatable closed-loop simulation for end-to-end and vision-language-action driving policies, but production deployment exposes three…
SpeechGym: An Audio-Native Gym for Training Voice Agents via Reinforcement Learning
Jiajun Fan, Jingyuan Li, Prashanth Gurunath Shivakumar +6
Voice agents must call tools and hold multi-turn dialogue entirely through speech, yet the dominant paradigm trains them in text. Existing frameworks either cascade TTS and ASR aro…
Can We Read the Mind of an Audio LLM? A Verbalizable, Multilingual Middle-Layer Workspace
Jiajun Fan, Jingyuan Li, Prashanth Gurunath Shivakumar +8
An audio language model is a black box in a specific way: we see what it says, never what it works out on the way there, and chain-of-thought monitoring helps only if the model wri…
Constraint-First Reasoning: A Training-Free Protocol for Exploiting Answer-Space Constraints in Mathematical Problem Solving
Hongbo Ma, Bangji Yang, Yunqian Selina Cheng +3
Large language models can derive a plausible mathematical object yet still violate explicit requirements--for example, by omitting a modular reduction, returning a non-integer, or…
Variable-Length Generative Protein Design via Generalized Poisson Flow
Chaoran Cheng, Zhanghan Ni, Yanru Qu +4
The ability to generate variable-length proteins is crucial in protein design, where the optimal length is often unknown and tightly coupled to designability. Current diffusion- an…
ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models
Ye Li, Huanan Liu, Kangye Ji +7
Vision-Language-Action (VLA) models are a powerful paradigm for generalist robotic control. However, their high computational cost and limited control frequency hinder real-time ro…
Retrieval is Cheap, Show Me the Code: Executable Multi-Hop Reasoning for Retrieval-Augmented Generation
Jiashuo Sun, Jimeng Shi, Yixuan Xie +10
Retrieval-Augmented Generation (RAG) has become a standard approach for knowledge-intensive question answering, but existing systems remain brittle on multi-hop questions, where so…
Trust or Abstain? A Self-Aware RAG Approach
Xi Zhu, Ziqi Wang, Kai Mei +5
Retrieval-augmented generation (RAG) improves large language models (LLMs) by incorporating external evidence, but it also introduces knowledge conflicts when retrieved contextual…
Batched Contextual Reinforcement: A Task-Scaling Law for Efficient Reasoning
Bangji Yang, Hongbo Ma, Jiajun Fan +1
Large Language Models employing Chain-of-Thought reasoning achieve strong performance but suffer from excessive token consumption that inflates inference costs. Existing efficiency…
M3: High-fidelity Text-to-Image Generation via Multi-Modal, Multi-Agent and Multi-Round Visual Reasoning
Bangji Yang, Ruihan Guo, Jiajun Fan +2
Generative models have achieved impressive fidelity in text-to-image synthesis, yet struggle with complex compositional prompts involving multiple constraints. We introduce \textbf…
Rethinking the Reranker: Boundary-Aware Evidence Selection for Robust Retrieval-Augmented Generation
Jiashuo Sun, Pengcheng Jiang, Saizhuo Wang +13
Retrieval-Augmented Generation (RAG) systems remain brittle under realistic retrieval noise, even when the required evidence appears in the top-K results. A key reason is that retr…
Incentivizing Consistent, Effective and Scalable Reasoning Capability in Audio LLMs via Reasoning Process Rewards
Jiajun Fan, Roger Ren, Jingyuan Li +6
The role of reasoning in Audio Large Language Models remains widely underexplored, as introducing a reasoning process often degrades rather than improves performance during inferen…
Fine-tuning Flow Matching Generative Models with Intermediate Feedback
Jiajun Fan, Chaoran Cheng, Shuaike Shen +2
Flow-based generative models have shown remarkable success in text-to-image generation, yet fine-tuning them with intermediate feedback remains challenging, especially for continuo…
Adaptive Divergence Regularized Policy Optimization for Fine-tuning Generative Models
Jiajun Fan, Tong Wei, Chaoran Cheng +2
Balancing exploration and exploitation during reinforcement learning fine-tuning of generative models presents a critical challenge, as existing approaches rely on fixed divergence…
ProteinZero: Self-Improving Protein Generation via Online Reinforcement Learning
Ziwen Wang, Jiajun Fan, Ruihan Guo +3
Protein generative models have shown remarkable promise in protein design, yet their success rates remain constrained by reliance on curated sequence-structure datasets and by misa…
Variational Supervised Contrastive Learning
Ziwen Wang, Jiajun Fan, Thao Nguyen +2
Contrastive learning has proven to be highly efficient and adaptable in shaping representation spaces across diverse modalities by pulling similar samples together and pushing diss…
-Flow: A Unified Framework for Continuous-State Discrete Flow Matching Models
Chaoran Cheng, Jiahan Li, Jiajun Fan +1
Recent efforts have extended the flow-matching framework to discrete generative modeling. One strand of models directly works with the continuous probabilities instead of discrete…
Online Reward-Weighted Fine-Tuning of Flow Matching with Wasserstein Regularization
Jiajun Fan, Shuaike Shen, Chaoran Cheng +3
Recent advancements in reinforcement learning (RL) have achieved great success in fine-tuning diffusion-based generative models. However, fine-tuning continuous flow-based generati…