95 citations · 181 across the 23 of their papers we have counts for
28 papers
Mind the Tail: Data Selection Through Sparse Concepts
Ziheng Zhang, Neha Hulkund, Colin Lee +7
Data selection reallocates training support across visual concepts, including objects, attributes, color, texture, and backgrounds. Different allocation strategies may produce vari…
Autonomous Driving Research Requires a Community-Driven Data Paradigm
Jinsu Yoo, Zanming Huang, Katie Z Luo +5
Autonomous driving has made remarkable progress, with recent AI advances enabling commercial deployments that are reshaping urban mobility. Yet the field remains far from its unive…
Agentic Context Learning with Self-Discovered Specification
Jike Zhong, Ming Li, Yuxiang Lai +8
Context learning is an emerging inference-time task where LLMs must learn and apply novel, task-specific knowledge from intricate contexts absent from pre-training; even frontier m…
A Study of Failure Modes in Two-Stage Human-Object Interaction Detection
Lemeng Wang, Qinqian Lei, Vidhi Bakshi +8
Human-object interaction (HOI) detection aims to detect interactions between humans and objects in images. While recent advances have improved performance on existing benchmarks, t…
Lessons and Open Questions from a Unified Study of Camera-Trap Species Recognition Over Time
Sooyoung Jeon, Hongjie Tian, Lemeng Wang +7
Camera traps are vital for large-scale biodiversity monitoring, yet accurate automated analysis remains challenging due to diverse deployment environments. While the computer visio…
Revisiting Model Stitching In the Foundation Model Era
Zheda Mai, Ke Zhang, Fu-En Wang +6
Model stitching, connecting early layers of one model (source) to later layers of another (target) via a light stitch layer, has served as a probe of representational compatibility…
Continual Unlearning for Text-to-Image Diffusion Models: A Regularization Perspective
Justin Lee, Zheda Mai, Jinsu Yoo +3
Machine unlearning--the ability to remove designated concepts from a pre-trained model--has advanced rapidly, particularly for text-to-image diffusion models. However, existing met…
AVA-Bench: Atomic Visual Ability Benchmark for Vision Foundation Models
Zheda Mai, Arpita Chowdhury, Zihe Wang +5
The rise of vision foundation models (VFMs) calls for systematic evaluation. A common approach pairs VFMs with large language models (LLMs) as general-purpose heads, followed by ev…
BioCLIP 2: Emergent Properties from Scaling Hierarchical Contrastive Learning
Jianyang Gu, Samuel Stevens, Elizabeth G Campolongo +13
Foundation models trained at scale exhibit remarkable emergent behaviors, learning new capabilities beyond their initial training objectives. We find such emergent behaviors in bio…
MEDA: Dynamic KV Cache Allocation for Efficient Multimodal Long-Context Inference
Zhongwei Wan, Hui Shen, Xin Wang +3
Long-context Multimodal Large Language Models (MLLMs) that incorporate long text-image and text-video modalities, demand substantial resources as their multimodal Key-Value (KV) ca…
Revisiting semi-supervised learning in the era of foundation models
Ping Zhang, Zheda Mai, Quang-Huy Nguyen +1
Semi-supervised learning (SSL) leverages abundant unlabeled data alongside limited labeled data to enhance learning. As vision foundation models (VFMs) increasingly serve as the ba…
Finer-CAM: Spotting the Difference Reveals Finer Details for Visual Explanation
Ziheng Zhang, Jianyang Gu, Arpita Chowdhury +5
Class activation map (CAM) has been widely used to highlight image regions that contribute to class predictions. Despite its simplicity and computational efficiency, CAM often stru…
Prompt-CAM: Making Vision Transformers Interpretable for Fine-Grained Analysis
Arpita Chowdhury, Dipanjyoti Paul, Zheda Mai +10
We present a simple approach to make pre-trained Vision Transformers (ViTs) interpretable for fine-grained analysis, aiming to identify and localize the traits that distinguish vis…
Fine-Tuning is Fine, if Calibrated
Zheda Mai, Arpita Chowdhury, Ping Zhang +8
Fine-tuning is arguably the most straightforward way to tailor a pre-trained model (e.g., a foundation model) to downstream applications, but it also comes with the risk of losing…
Lessons and Insights from a Unifying Study of Parameter-Efficient Fine-Tuning (PEFT) in Visual Recognition
Zheda Mai, Ping Zhang, Cheng-Hao Tu +3
Parameter-efficient fine-tuning (PEFT) has attracted significant attention due to the growth of pre-trained model sizes and the need to fine-tune (FT) them for superior downstream…
MLLM-CompBench: A Comparative Reasoning Benchmark for Multimodal LLMs
Jihyung Kil, Zheda Mai, Justin Lee +6
The ability to compare objects, scenes, or situations is crucial for effective decision-making and problem-solving in everyday life. For instance, comparing the freshness of apples…
Holistic Transfer: Towards Non-Disruptive Fine-Tuning with Partial Target Data
Cheng-Hao Tu, Hong-You Chen, Zheda Mai +7
We propose a learning problem involving adapting a pre-trained source model to the target domain for classifying all classes that appeared in the source data, using target data tha…
Segment Anything Model (SAM) Enhanced Pseudo Labels for Weakly Supervised Semantic Segmentation
Tianle Chen, Zheda Mai, Ruiwen Li +1
Weakly supervised semantic segmentation (WSSS) aims to bypass the need for laborious pixel-level annotation by using only image-level annotation. Most existing methods rely on Clas…
Visual Query Tuning: Towards Effective Usage of Intermediate Representations for Parameter and Memory Efficient Transfer Learning
Cheng-Hao Tu, Zheda Mai, Wei-Lun Chao
Intermediate features of a pre-trained model have been shown informative for making accurate predictions on downstream tasks, even if the model backbone is kept frozen. The key cha…
TransCAM: Transformer Attention-based CAM Refinement for Weakly Supervised Semantic Segmentation
Ruiwen Li, Zheda Mai, Chiheb Trabelsi +3
Weakly supervised semantic segmentation (WSSS) with only image-level supervision is a challenging task. Most existing methods exploit Class Activation Maps (CAM) to generate pixel-…
Unintended Bias in Language Model-driven Conversational Recommendation
Tianshu Shen, Jiaru Li, Mohamed Reda Bouadjenek +2
Conversational Recommendation Systems (CRSs) have recently started to leverage pretrained language models (LM) such as BERT for their ability to semantically interpret a wide range…
Supervised Contrastive Replay: Revisiting the Nearest Class Mean Classifier in Online Class-Incremental Continual Learning
Zheda Mai, Ruiwen Li, Hyunwoo Kim +1
Online class-incremental continual learning (CL) studies the problem of learning new classes continually from an online non-stationary data stream, intending to adapt to new data w…
Online Continual Learning in Image Classification: An Empirical Survey
Zheda Mai, Ruiwen Li, Jihwan Jeong +3
Online continual learning for image classification studies the problem of learning to classify images from an online stream of data and tasks, where tasks may include new classes (…
Attentive Autoencoders for Multifaceted Preference Learning in One-class Collaborative Filtering
Zheda Mai, Ga Wu, Kai Luo +1
Most existing One-Class Collaborative Filtering (OC-CF) algorithms estimate a user's preference as a latent vector by encoding their historical interactions. However, users often s…