中文
相关论文

相关论文: SymDPO: Boosting In-Context Learning of Large Mult…

200 篇论文

Large Language Models (LLMs) have demonstrated remarkable potential in automating software development tasks. While recent advances leverage Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) to align models with human…

软件工程 · 计算机科学 2025-12-09 Xin Yin , Chao Ni , Xiaohu Yang

A common technique for aligning large language models (LLMs) relies on acquiring human preferences by comparing multiple generations conditioned on a fixed context. This method, however, relies solely on pairwise comparisons, where the…

计算与语言 · 计算机科学 2025-01-09 Hritik Bansal , Ashima Suvarna , Gantavya Bhatt , Nanyun Peng , Kai-Wei Chang , Aditya Grover

The emergence of large Vision Language Models (VLMs) has broadened the scope and capabilities of single-modal Large Language Models (LLMs) by integrating visual modalities, thereby unlocking transformative cross-modal applications in a…

计算机视觉与模式识别 · 计算机科学 2025-09-23 Shuo Xing , Peiran Li , Yuping Wang , Ruizheng Bai , Yueqi Wang , Chan-Wei Hu , Chengxuan Qian , Huaxiu Yao , Zhengzhong Tu

Most Video Large Language Models (Video-LLMs) adopt preference alignment techniques, e.g., DPO~\citep{rafailov2024dpo}, to optimize the reward margin between a winning response ($y_w$) and a losing response ($y_l$). However, the likelihood…

计算机视觉与模式识别 · 计算机科学 2025-06-06 Xiaodong Wang , Jinfa Huang , Li Yuan , Peixi Peng

As large language models (LLMs) are increasingly applied across various domains, enhancing safety while maintaining the helpfulness of LLMs has become a critical challenge. Recent studies solve this problem through safety-constrained online…

计算与语言 · 计算机科学 2025-06-04 Yupeng Qi , Ziyu Lyu , Min Yang , Yanlin Wang , Lu Bai , Lixin Cui

Vision generation remains a challenging frontier in artificial intelligence, requiring seamless integration of visual understanding and generative capabilities. In this paper, we propose a novel framework, Vision-Driven Prompt Optimization…

计算机视觉与模式识别 · 计算机科学 2025-01-07 Leo Franklin , Apiradee Boonmee , Kritsada Wongsuwan

As development of large language models (LLM) progresses, aligning them with human preferences has become increasingly important. We propose stepwise DPO (sDPO), an extension of the recently popularized direct preference optimization (DPO)…

计算与语言 · 计算机科学 2024-10-08 Dahyun Kim , Yungi Kim , Wonho Song , Hyeonwoo Kim , Yunsu Kim , Sanghoon Kim , Chanjun Park

Large Language Models (LLMs) have become increasingly popular due to their ability to process and generate natural language. However, as they are trained on massive datasets of text, LLMs can inherit harmful biases and produce outputs that…

计算与语言 · 计算机科学 2025-01-23 Qi Gou , Cam-Tu Nguyen

Existing open-source multimodal large language models (MLLMs) generally follow a training process involving pre-training and supervised fine-tuning. However, these models suffer from distribution shifts, which limit their multimodal…

计算与语言 · 计算机科学 2025-04-08 Weiyun Wang , Zhe Chen , Wenhai Wang , Yue Cao , Yangzhou Liu , Zhangwei Gao , Jinguo Zhu , Xizhou Zhu , Lewei Lu , Yu Qiao , Jifeng Dai

Fine-grained video captioning aims to generate detailed, temporally coherent descriptions of video content. However, existing methods struggle to capture subtle video dynamics and rich detailed information. In this paper, we leverage…

人工智能 · 计算机科学 2026-03-24 Jisheng Dang , Yizhou Zhang , Hao Ye , Teng Wang , Siming Chen , Huicheng Zheng , Yulan Guo , Jianhuang Lai , Bin Hu

Large Video Models (LVMs) built upon Large Language Models (LLMs) have shown promise in video understanding but often suffer from misalignment with human intuition and video hallucination issues. To address these challenges, we introduce…

计算机视觉与模式识别 · 计算机科学 2025-04-18 Haojian Huang , Haodong Chen , Shengqiong Wu , Meng Luo , Jinlan Fu , Xinya Du , Hanwang Zhang , Hao Fei

With the rapid development and widespread application of Large Language Models (LLMs), their potential safety risks have attracted widespread attention. Reinforcement Learning from Human Feedback (RLHF) has been adopted to enhance the…

Masked diffusion large language models (dLLMs) are emerging as promising alternatives to autoregressive LLMs, offering competitive performance while supporting unique generation capabilities such as inpainting. We explore how inpainting can…

机器学习 · 计算机科学 2025-09-15 Siyan Zhao , Mengchen Liu , Jing Huang , Miao Liu , Chenyu Wang , Bo Liu , Yuandong Tian , Guan Pang , Sean Bell , Aditya Grover , Feiyu Chen

Existing multi-objective preference alignment methods for large language models (LLMs) face limitations: (1) the inability to effectively balance various preference dimensions, and (2) reliance on auxiliary reward/reference models…

机器学习 · 计算机科学 2025-06-10 Qi Liu , Jingqing Ruan , Hao Li , Haodong Zhao , Desheng Wang , Jiansong Chen , Wan Guanglu , Xunliang Cai , Zhi Zheng , Tong Xu

Aligning Large Language Models (LLMs) with human preferences is critical, yet traditional fine-tuning methods are computationally expensive and inflexible. While test-time alignment offers a promising alternative, existing approaches often…

人工智能 · 计算机科学 2026-01-16 Tiesunlong Shen , Rui Mao , Jin Wang , Heming Sun , Jian Zhang , Xuejie Zhang , Erik Cambria

Direct Preference Optimization (DPO) has become a widely used training method for the instruction fine-tuning of large language models (LLMs). In this work, we explore an under-investigated aspect of DPO - its dependency on the reference…

计算与语言 · 计算机科学 2024-08-23 Yixin Liu , Pengfei Liu , Arman Cohan

Large language models (LLMs) famously exhibit emergent in-context learning (ICL) -- the ability to rapidly adapt to new tasks using few-shot examples provided as a prompt, without updating the model's weights. Built on top of LLMs, vision…

机器学习 · 计算机科学 2025-04-02 Yongshuo Zong , Ondrej Bohdal , Timothy Hospedales

Generative AI can now synthesize strikingly realistic images from text, yet output quality remains highly sensitive to how prompts are phrased. Direct Preference Optimization (DPO) offers a lightweight, off-policy alternative to RL for…

计算与语言 · 计算机科学 2025-07-30 Anas Mohamed , Azal Ahmad Khan , Xinran Wang , Ahmad Faraz Khan , Shuwen Ge , Saman Bahzad Khan , Ayaan Ahmad , Ali Anwar

In-context learning (ICL) is a new paradigm for natural language processing (NLP), where a large language model (LLM) observes a small number of demonstrations and a test instance as its input, and directly makes predictions without…

计算与语言 · 计算机科学 2024-12-12 Fei Zhao , Taotian Pang , Zhen Wu , Zheng Ma , Shujian Huang , Xinyu Dai

Through alignment with human preferences, Large Language Models (LLMs) have advanced significantly in generating honest, harmless, and helpful responses. However, collecting high-quality preference data is a resource-intensive and…

计算与语言 · 计算机科学 2024-10-10 Qingxiu Dong , Li Dong , Xingxing Zhang , Zhifang Sui , Furu Wei