67 citations · 81 across the 9 of their papers we have counts for
9 papers
GRRR: The Geometry of Reshaping, Rotation, and Routing in Decoder LLM post-training
Jianing Qi, Hao Tang, Zhigang Zhu
We study how post-training changes the weights of Large Language Models (LLMs) relative to their pretrained weights. Across 12 post-training chains with supervised fine-tuning (SFT…
Policy Gradient Guidance Enables Test Time Control
Jianing Qi, Hao Tang, Zhigang Zhu
We introduce Policy Gradient Guidance (PGG), a simple extension of classifier-free guidance from diffusion models to classical policy gradient methods. PGG augments the policy grad…
Learning to Reason Across Parallel Samples for LLM Reasoning
Jianing Qi, Xi Ye, Hao Tang +2
Scaling test-time compute brings substantial performance gains for large language models (LLMs). By sampling multiple answers and heuristically aggregate their answers (e.g., eithe…
Beyond Semantics: Rediscovering Spatial Awareness in Vision-Language Models
Jianing Qi, Jiawei Liu, Hao Tang +1
Vision Language Models (VLMs) excel at identifying and describing objects but often fail at spatial reasoning. We study why VLMs, such as LLaVA, underutilize spatial cues despite h…
VerifierQ: Enhancing LLM Test Time Compute with Q-Learning-based Verifiers
Jianing Qi, Hao Tang, Zhigang Zhu
Recent advancements in test time compute, particularly through the use of verifier models, have significantly enhanced the reasoning capabilities of Large Language Models (LLMs). T…
GMC: A General Framework of Multi-stage Context Learning and Utilization for Visual Detection Tasks
Xuan Wang, Hao Tang, Zhigang Zhu
Various contextual information has been employed by many approaches for visual detection tasks. However, most of the existing approaches only focus on specific context for specific…
Pedestrian Accessible Infrastructure Inventory: Assessing Zero-Shot Segmentation on Multi-Mode Geospatial Data for All Pedestrian Types
Jiahao Xia, Gavin Gong, Jiawei Liu +2
In this paper, a Segment Anything Model (SAM)-based pedestrian infrastructure segmentation workflow is designed and optimized, which is capable of efficiently processing multi-sour…
Improving Dense Crowd Counting Convolutional Neural Networks using Inverse k-Nearest Neighbor Maps and Multiscale Upsampling
Greg Olmschenk, Hao Tang, Zhigang Zhu
Gatherings of thousands to millions of people frequently occur for an enormous variety of events, and automated counting of these high-density crowds is useful for safety, manageme…
Generalizing semi-supervised generative adversarial networks to regression using feature contrasting
Greg Olmschenk, Zhigang Zhu, Hao Tang
In this work, we generalize semi-supervised generative adversarial networks (GANs) from classification problems to regression problems. In the last few years, the importance of imp…