5 citations · 14 across the 21 of their papers we have counts for
22 papers
VPG: Visual Prefix Guidance for Autoregressive Image and Video Generation
Xinyao Liao, Qiyuan He, Yicong Li +4
Autoregressive image and video generators are trained with teacher-forced histories but must sample from their own generated prefixes at inference time, making them vulnerable to e…
Stable-DiffCoder: Pushing the Frontier of Code Diffusion Large Language Model
Chenghao Fan, Wen Heng, Bo Li +6
Diffusion-based language models (DLLMs) offer non-sequential, block-wise generation and richer data reuse compared to autoregressive (AR) models, but existing code DLLMs still lag…
VA-: Variational Policy Alignment for Pixel-Aware Autoregressive Generation
Xinyao Liao, Qiyuan He, Kai Xu +4
Autoregressive (AR) visual generation relies on tokenizers to map images to and from discrete sequences. However, tokenizers are trained to reconstruct clean images from ground-tru…
Scaling Reasoning, Losing Control: Evaluating Instruction Following in Large Reasoning Models
Tingchen Fu, Jiawei Gu, Yafu Li +2
Instruction-following is essential for aligning large language models (LLMs) with user intent. While recent reasoning-oriented models exhibit impressive performance on complex math…
CoCA: Step-level Reward for Free in RL-based T2I Diffusion Model Fine-tuning
Xinyao Liao, Wei Wei, Xiaoye Qu +3
Recent advances in text-to-image (T2I) diffusion model fine-tuning leverage reinforcement learning (RL) to align generated images with learnable reward functions. The existing appr…
SATORI-R1: Incentivizing Multimodal Reasoning through Explicit Visual Anchoring
Chuming Shen, Wei Wei, Xiaoye Qu +1
DeepSeek-R1 has demonstrated powerful reasoning capabilities in the text domain through stable reinforcement learning (RL). Recently, in the multimodal domain, works have begun to…
Extrapolating and Decoupling Image-to-Video Generation Models: Motion Modeling is Easier Than You Think
Jie Tian, Xiaoye Qu, Zhenyi Lu +3
Image-to-Video (I2V) generation aims to synthesize a video clip according to a given image and condition (e.g., text). The key challenge of this task lies in simultaneously generat…
A Survey of Efficient Reasoning for Large Reasoning Models: Language, Multimodality, and Beyond
Xiaoye Qu, Yafu Li, Zhao-Chen Su +15
Recent Large Reasoning Models (LRMs), such as DeepSeek-R1 and OpenAI o1, have demonstrated strong performance gains by scaling up the length of Chain-of-Thought (CoT) reasoning dur…
Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment
Chenghao Fan, Zhenyi Lu, Sichen Liu +4
While Low-Rank Adaptation (LoRA) enables parameter-efficient fine-tuning for Large Language Models (LLMs), its performance often falls short of Full Fine-Tuning (Full FT). Current…
SURf: Teaching Large Vision-Language Models to Selectively Utilize Retrieved Information
Jiashuo Sun, Jihai Zhang, Yucheng Zhou +3
Large Vision-Language Models (LVLMs) have become pivotal at the intersection of computer vision and natural language processing. However, the full potential of LVLMs Retrieval-Augm…
Look, Compare, Decide: Alleviating Hallucination in Large Vision-Language Models via Multi-View Multi-Path Reasoning
Xiaoye Qu, Jiashuo Sun, Wei Wei +1
Recently, Large Vision-Language Models (LVLMs) have demonstrated impressive capabilities in multi-modal context comprehension. However, they still suffer from hallucination problem…
Alleviating Hallucination in Large Vision-Language Models with Active Retrieval Augmentation
Xiaoye Qu, Qiyuan Chen, Wei Wei +2
Despite the remarkable ability of large vision-language models (LVLMs) in image comprehension, these models frequently generate plausible yet factually incorrect responses, a pheno…
Mitigating Multilingual Hallucination in Large Vision-Language Models
Xiaoye Qu, Mingyang Song, Wei Wei +2
While Large Vision-Language Models (LVLMs) have exhibited remarkable capabilities across a wide range of tasks, they suffer from hallucination problems, where models generate plaus…
LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training
Tong Zhu, Xiaoye Qu, Daize Dong +4
Mixture-of-Experts (MoE) has gained increasing popularity as a promising framework for scaling up large language models (LLMs). However, training MoE from scratch in a large-scale…
Mitigating Boundary Ambiguity and Inherent Bias for Text Classification in the Era of Large Language Models
Zhenyi Lu, Jie Tian, Wei Wei +4
Text classification is a crucial task encountered frequently in practical scenarios, yet it is still under-explored in the era of large language models (LLMs). This study shows tha…
Improving Pseudo Labels with Global-Local Denoising Framework for Cross-lingual Named Entity Recognition
Zhuojun Ding, Wei Wei, Xiaoye Qu +1
Cross-lingual named entity recognition (NER) aims to train an NER model for the target language leveraging only labeled source language data and unlabeled target language data. Pri…
On Giant's Shoulders: Effortless Weak to Strong by Dynamic Logits Fusion
Chenghao Fan, Zhenyi Lu, Wei Wei +4
Efficient fine-tuning of large language models for task-specific applications is imperative, yet the vast number of parameters in these models makes their training increasingly cha…
Twin-Merging: Dynamic Integration of Modular Expertise in Model Merging
Zhenyi Lu, Chenghao Fan, Wei Wei +3
In the era of large language models, model merging is a promising way to combine multiple task-specific models into a single multitask model without extra training. However, two ch…
Learning the Unlearned: Mitigating Feature Suppression in Contrastive Learning
Jihai Zhang, Xiang Lan, Xiaoye Qu +3
Self-Supervised Contrastive Learning has proven effective in deriving high-quality representations from unlabeled data. However, a major challenge that hinders both unimodal and mu…
Joint Multi-Facts Reasoning Network For Complex Temporal Question Answering Over Knowledge Graph
Rikui Huang, Wei Wei, Xiaoye Qu +3
Temporal Knowledge Graph (TKG) is an extension of regular knowledge graph by attaching the time scope. Existing temporal knowledge graph question answering (TKGQA) models solely ap…
Enhancing Low-Resource Relation Representations through Multi-View Decoupling
Chenghao Fan, Wei Wei, Xiaoye Qu +4
Recently, prompt-tuning with pre-trained language models (PLMs) has demonstrated the significantly enhancing ability of relation extraction (RE) tasks. However, in low-resource sce…
MIRACLE: Towards Personalized Dialogue Generation with Latent-Space Multiple Personal Attribute Control
Zhenyi Lu, Wei Wei, Xiaoye Qu +3
Personalized dialogue systems aim to endow the chatbot agent with more anthropomorphic traits for human-like interactions. Previous approaches have explored explicitly user profile…