12 papers
Learning from Failures: Retrieval-Centric CoT via Hard Negatives for Unified Multimodal Retrieval
Zelong Sun, Jun Wang, Kaicheng Yang +3
Unified multimodal retrieval aims to identify candidates that satisfy complex user intent expressed through heterogeneous inputs. Although Large Vision-Language Model (LVLM)-based…
Memento 2: Learning by Stateful Reflective Memory
Jun Wang
We present a theoretical study of continual and experiential learning in large language model agents that combine episodic memory with reinforcement learning. We argue that the key…
Can Synthetic Images Serve as Effective and Efficient Class Prototypes?
Dianxing Shi, Dingjie Fu, Yuqiao Liu +1
Vision-Language Models (VLMs) have shown strong performance in zero-shot image classification tasks. However, existing methods, including Contrastive Language-Image Pre-training (C…
DanQing: An Up-to-Date Large-Scale Chinese Vision-Language Pre-training Dataset
Hengyu Shen, Tiancheng Gu, Bin Qin +10
Vision-Language Pre-training (VLP) models have achieved remarkable success by leveraging large-scale image-text pairs. While English-centric models like CLIP and SigLIP benefit fro…
OmniInfer: System-Wide Acceleration Techniques for Optimizing LLM Serving Throughput and Latency
Jun Wang, Yunxiang Yao, Wenwei Kuang +11
Large Language Models drive a wide range of modern AI applications but impose substantial challenges on large-scale serving systems due to intensive computation, strict latency con…
AMLA: MUL by ADD in FlashAttention Rescaling
Qichen Liao, Chengqiu Hu, Fangzheng Miao +8
Multi-head Latent Attention (MLA) significantly reduces KVCache memory usage in Large Language Models while introducing substantial computational overhead and intermediate variable…
Dynamic Pattern Alignment Learning for Pretraining Lightweight Human-Centric Vision Models
Xuanhan Wang, Huimin Deng, Ke Liu +3
Human-centric vision models (HVMs) have achieved remarkable generalization due to large-scale pretraining on massive person images. However, their dependence on large neural archit…
ProVision: Programmatically Scaling Vision-centric Instruction Data for Multimodal Language Models
Jieyu Zhang, Le Xue, Linxin Song +11
With the rise of multimodal applications, instruction data has become critical for training multimodal language models capable of understanding complex image-based queries. Existin…
BLIP3-KALE: Knowledge Augmented Large-Scale Dense Captions
Anas Awadalla, Le Xue, Manli Shu +13
We introduce BLIP3-KALE, a dataset of 218 million image-text pairs that bridges the gap between descriptive synthetic captions and factual web-scale alt-text. KALE augments synthet…
Elucidating the design space of language models for image generation
Xuantong Liu, Shaozhe Hao, Xianbiao Qi +4
The success of autoregressive (AR) language models in text generation has inspired the computer vision community to adopt Large Language Models (LLMs) for image generation. However…
UFT: Unifying Fine-Tuning of SFT and RLHF/DPO/UNA through a Generalized Implicit Reward Function
Zhichao Wang, Bin Bi, Zixu Zhu +6
By pretraining on trillions of tokens, an LLM gains the capability of text generation. However, to enhance its utility and reduce potential harm, SFT and alignment are applied sequ…
Don't Throw Away Data: Better Sequence Knowledge Distillation
Jun Wang, Eleftheria Briakou, Hamid Dadkhahi +3
A critical component in knowledge distillation is the means of coupling the teacher and student. The predominant sequence knowledge distillation method involves supervised learning…