10 papers
Uncovering and Mitigating Aggregation-Induced Reward Hacking in Multi-Reward Reinforcement Learning
Yu Yuan, Yaoyou Fan, Lili Zhao +5
Reinforcement learning fine-tuning of large language models increasingly adopts multiple reward dimensions, including verifiable rules, task-specific evaluators, and learned reward…
GenFirst: Generation Before Reconstruction for Stable End-to-End Latent Generative Modeling
Guangting Zheng, Yiyuan Zhang, Tao Yang +4
Latent generative models typically follow a two-stage pipeline, training a variational autoencoder for reconstruction and then a generative model on the frozen latent space. Since…
RoMAN-Flow: Taming Autoregressive Normalizing Flows for Offline Reinforcement Learning in Robotic Manipulation
Shaoxuan Wang, Guangting Zheng, Rui Huang +4
Offline reinforcement learning improves robotic policies using previously collected data without further environment interaction. Yet prevalent diffusion- and flow-matching robot p…
Enhancing Video Physical Consistency via Role-aware Joint Training and Modality-decoupled Denoising
Guangting Zheng, Haojing Chen, Hao Li +6
While modern video diffusion models excel in visual fidelity, maintaining long-range physical consistency remains a formidable challenge. Conventional pixel-reconstruction objectiv…
SimFlow: Simplified and End-to-End Training of Latent Normalizing Flows
Qinyu Zhao, Guangting Zheng, Tao Yang +4
Normalizing Flows (NFs) learn invertible mappings between the data and a Gaussian distribution. Prior works usually suffer from two limitations. First, they add random noise to tra…
FARMER: Flow AutoRegressive Transformer over Pixels
Guangting Zheng, Qinyu Zhao, Tao Yang +6
Directly modeling the explicit likelihood of the raw data distribution is key topic in the machine learning area, which achieves the scaling successes in Large Language Models by a…
Simulating Human-Like Learning Dynamics with LLM-Empowered Agents
Yu Yuan, Lili Zhao, Wei Chen +4
Capturing human learning behavior based on deep learning methods has become a major research focus in both psychology and intelligent systems. Recent approaches rely on controlled…
Hierarchical Masked Autoregressive Models with Low-Resolution Token Pivots
Guangting Zheng, Yehao Li, Yingwei Pan +4
Autoregressive models have emerged as a powerful generative paradigm for visual generation. The current de-facto standard of next token prediction commonly operates over a single-s…
S3R-GS: Streamlining the Pipeline for Large-Scale Street Scene Reconstruction
Guangting Zheng, Jiajun Deng, Xiaomeng Chu +3
Recently, 3D Gaussian Splatting (3DGS) has reshaped the field of photorealistic 3D reconstruction, achieving impressive rendering quality and speed. However, when applied to large-…
Do LLMs Overcome Shortcut Learning? An Evaluation of Shortcut Challenges in Large Language Models
Yu Yuan, Lili Zhao, Kai Zhang +2
Large Language Models (LLMs) have shown remarkable capabilities in various natural language processing tasks. However, LLMs may rely on dataset biases as shortcuts for prediction,…