17 papers
GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed
Qize Yu, Lianrui Fan, Bowen Ping +19
Autoregressive (AR) grounding models serialize spatial predictions, introducing sequential latency and imposing a causal order on output tokens. We view grounding as visual evidenc…
AffordanceWAM: Affordance-Aware Joint World-Action Modeling for Robot Manipulation
Jiadi You, Qize Yu, Yue Chen +11
Generalizable robot manipulation requires predicting how a scene will evolve, identifying where interactions are feasible, and determining how to act. Action-labeled robot videos d…
CameraEditor: Camera-Controlled Image Editing via Video-Prior Sequential Modeling
Xin Shen, Chengyou Jia, Keshuo Xing +6
Beyond semantic content, camera parameters play a pivotal role in dictating the geometric perspective and appearance of any given image. While recent image editing models excel at…
Predictive Divergence Masks for LLM RL
Xiangxin Zhou, Jiarui Yao, Penghui Qi +4
Reinforcement learning for large language models (LLMs) typically relies on trust-region masks to stabilize off-policy updates. The dominant PPO-style approach uses the sampled-tok…
Flow-DPPO: Divergence Proximal Policy Optimization for Flow Matching Models
Bowen Ping, Xiangxin Zhou, Penghui Qi +3
Recent work has demonstrated that online reinforcement learning (RL) can substantially improve the quality and alignment of flow matching models for image and video generation. Met…
AffordanceVLA: A Vision-Language-Action Model Empowering Action Generation through Affordance-Aware Understanding
Qize Yu, Jiadi You, Yuran Wang +12
Vision-Language-Action (VLA) models leverage the rich world knowledge of pretrained vision-language models (VLMs) to enable instruction-following robotic manipulation. However, the…
OmniTryOn: Video Try-On Anything at Once!
Changliang Xia, Chengyou Jia, Minnan Luo +3
Although video virtual try-on (VVT) has achieved significant progress, existing methods still exhibit two fundamental limitations: first, they are restricted to single-garment tran…
LongAct: Harnessing Intrinsic Activation Patterns for Long-Context Reinforcement Learning
Bowen Ping, Zijun Chen, Tingfeng Hui +4
Reinforcement Learning (RL) has emerged as a critical driver for enhancing the reasoning capabilities of Large Language Models (LLMs). While recent advancements have focused on rew…
Flow-Factory: A Unified Framework for Reinforcement Learning in Flow-Matching Models
Bowen Ping, Chengyou Jia, Minnan Luo +2
Reinforcement learning has emerged as a promising paradigm for aligning diffusion and flow-matching models with human preferences, yet practitioners face fragmented codebases, mode…
LongR: Unleashing Long-Context Reasoning via Reinforcement Learning with Dense Utility Rewards
Qize Yu, Bowen Ping, Zijun Chen +6
Reinforcement Learning (RL) has emerged as a key driver for LLM reasoning. This capability is equally pivotal in long-context scenarios---such as long-dialogue understanding and st…
-Predictor: Noise-Free Deterministic Diffusion for Dense Prediction
Changliang Xia, Chengyou Jia, Minnan Luo +3
Although diffusion models with strong visual priors have emerged as powerful dense prediction backbones, they overlook a core limitation: the stochastic noise at the core of diffus…
PaCo-RL: Advancing Reinforcement Learning for Consistent Image Generation with Pairwise Reward Modeling
Bowen Ping, Chengyou Jia, Minnan Luo +4
Consistent image generation requires faithfully preserving identities, styles, and logical coherence across multiple images, which is essential for applications such as storytellin…
DecIF: Improving Instruction-Following through Meta-Decomposition
Tingfeng Hui, Pengyu Zhu, Bowen Ping +4
Instruction-following has emerged as a crucial capability for large language models (LLMs). However, existing approaches often rely on pre-existing documents or external resources…
AutoGPS: Automated Geometry Problem Solving via Multimodal Formalization and Deductive Reasoning
Bowen Ping, Minnan Luo, Zhuohang Dang +2
Geometry problem solving presents distinctive challenges in artificial intelligence, requiring exceptional multimodal comprehension and rigorous mathematical reasoning capabilities…
LongDPO: Unlock Better Long-form Generation Abilities for LLMs via Critique-augmented Stepwise Information
Bowen Ping, Jiali Zeng, Fandong Meng +3
Long-form generation is crucial for academic writing papers and repo-level code generation. Despite this, current models, including GPT-4o, still exhibit unsatisfactory performance…
Delta-CoMe: Training-Free Delta-Compression with Mixed-Precision for Large Language Models
Bowen Ping, Shuo Wang, Hanqing Wang +7
Fine-tuning is a crucial process for adapting large language models (LLMs) to diverse applications. In certain scenarios, such as multi-tenant serving, deploying multiple LLMs beco…
LoRA-Flow: Dynamic LoRA Fusion for Large Language Models in Generative Tasks
Hanqing Wang, Bowen Ping, Shuo Wang +4
LoRA employs lightweight modules to customize large language models (LLMs) for each downstream task or domain, where different learned additional modules represent diverse skills.…