English
Related papers

Related papers: Keep the General, Inject the Specific: Structured …

200 papers

Recent trends in Large Vision Language Models (LVLMs) research have been increasingly focusing on advancing beyond general image understanding towards more nuanced, object-level referential comprehension. In this paper, we present and delve…

Computer Vision and Pattern Recognition · Computer Science 2024-03-21 Tongtian Yue , Jie Cheng , Longteng Guo , Xingyuan Dai , Zijia Zhao , Xingjian He , Gang Xiong , Yisheng Lv , Jing Liu

Large Language Models (LLMs) display strikingly different generalization behaviors: supervised fine-tuning (SFT) often narrows capability, whereas reinforcement-learning (RL) tuning tends to preserve it. The reasons behind this divergence…

Machine Learning · Computer Science 2026-01-01 Haoyue Bai , Yiyou Sun , Wenjie Hu , Shi Qiu , Maggie Ziyu Huan , Peiyang Song , Robert Nowak , Dawn Song

Neural dialog models have exhibited strong performance, however their end-to-end nature lacks a representation of the explicit structure of dialog. This results in a loss of generalizability, controllability and a data-hungry nature.…

Computation and Language · Computer Science 2019-07-24 Shikib Mehri , Tejas Srinivasan , Maxine Eskenazi

Alignment, endowing a pre-trained Large language model (LLM) with the ability to follow instructions, is crucial for its real-world applications. Conventional supervised fine-tuning (SFT) methods formalize it as causal language modeling…

Computation and Language · Computer Science 2024-12-18 Yuchen Fan , Yuzhong Hong , Qiushi Wang , Junwei Bao , Hongfei Jiang , Yang Song

Existing domain-specific Large Language Models (LLMs) are typically developed by fine-tuning general-purposed LLMs with large-scale domain-specific corpora. However, training on large-scale corpora often fails to effectively organize domain…

Computation and Language · Computer Science 2025-02-11 Zhiqiang Liu , Chengtao Gan , Junjie Wang , Yichi Zhang , Zhongpu Bo , Mengshu Sun , Huajun Chen , Wen Zhang

Continual learning in large language models (LLMs) is prone to catastrophic forgetting, where adapting to new tasks significantly degrades performance on previously learned ones. Existing methods typically rely on low-rank,…

We study the limitations of Large Language Models (LLMs) for the task of response generation in human-machine dialogue. Several techniques have been proposed in the literature for different dialogue types (e.g., Open-Domain). However, the…

Computation and Language · Computer Science 2024-08-06 Simone Alghisi , Massimo Rizzoli , Gabriel Roccabruna , Seyed Mahed Mousavi , Giuseppe Riccardi

In this work, we present the first systematic evaluation of catastrophic forgetting and modality inequivalence in speech large language models, showing that introducing speech capabilities can degrade knowledge and reasoning even when…

Computation and Language · Computer Science 2025-09-19 Enzhi Wang , Qicheng Li , Zhiyuan Tang , Yuhang Jia

Advancements in Large Language Models (LLMs) have significantly enhanced instruction-following capabilities. However, most Instruction Fine-Tuning (IFT) datasets are predominantly in English, limiting model performance in other languages.…

Computation and Language · Computer Science 2024-07-03 Sathish Reddy Indurthi , Wenxuan Zhou , Shamil Chollampatt , Ravi Agrawal , Kaiqiang Song , Lingxiao Zhao , Chenguang Zhu

Large Language Models (LLMs) demonstrate exceptional reasoning capabilities, often achieving state-of-the-art performance in various tasks. However, their substantial computational and memory demands, due to billions of parameters, hinder…

Computation and Language · Computer Science 2024-11-25 Xunyu Zhu , Jian Li , Can Ma , Weiping Wang

Large Language Models (LLMs) are increasingly employed in multi-turn conversational tasks, yet their pre-training data predominantly consists of continuous prose, creating a potential mismatch between required capabilities and training…

Computation and Language · Computer Science 2025-07-09 Jing Yang Lee , Hamed Bonab , Nasser Zalmout , Ming Zeng , Sanket Lokegaonkar , Colin Lockard , Binxuan Huang , Ritesh Sarkhel , Haodong Wang

Pre-trained Vision-Language Models (VLMs) require Continual Learning (CL) to efficiently update their knowledge and adapt to various downstream tasks without retraining from scratch. However, for VLMs, in addition to the loss of knowledge…

Computer Vision and Pattern Recognition · Computer Science 2025-03-07 Bin Wu , Wuxuan Shi , Jinqiao Wang , Mang Ye

This paper explores a scientific question in supervised fine-tuning (SFT): why SFT is broadly effective for small-scale deep neural networks, yet can produce inconsistent or even detrimental effects when applied to large language models…

Artificial Intelligence · Computer Science 2026-05-19 Junpeng Zhang , Lei Cheng , Guoxi Zhang , Hua Cai , Qing Xu , Quanshi Zhang

In traditional innovation practices, concept and IP generation are often iteratively integrated. Both processes demand an intricate understanding of advanced technical domain knowledge. Existing large language models (LLMs), while…

Computation and Language · Computer Science 2025-04-09 Runtao Ren , Jian Ma , Jianxi Luo

Unconstrained fine-tuning of flow-matching Vision-Language-Action (VLA) models drives dense parameter overwrites, degrading pre-trained capabilities. We present Conservative Supervised Fine-Tuning (ConSFT), an optimization objective that…

Robotics · Computer Science 2026-05-20 Tianyi Zhang , Shaopeng Zhai , Haoran Zhang , Fuxian Huang , Qi Zhang

The alignment of large language models (LLMs) is crucial not only for unlocking their potential in specific tasks but also for ensuring that responses meet human expectations and adhere to safety and ethical principles. Current alignment…

Computation and Language · Computer Science 2024-06-18 Ruijun Chen , Jiehao Liang , Shiping Gao , Fanqi Wan , Xiaojun Quan

Reasoning in mathematical domains remains a significant challenge for relatively small language models (LMs). Many current methods focus on specializing LMs in mathematical reasoning and rely heavily on knowledge distillation from powerful…

Artificial Intelligence · Computer Science 2023-07-18 Zhenwen Liang , Dian Yu , Xiaoman Pan , Wenlin Yao , Qingkai Zeng , Xiangliang Zhang , Dong Yu

Pre-trained neural language models bring significant improvement for various NLP tasks, by fine-tuning the models on task-specific training sets. During fine-tuning, the parameters are initialized from pre-trained models directly, which…

Computation and Language · Computer Science 2020-09-17 Chengyu Wang , Minghui Qiu , Jun Huang , Xiaofeng He

Supervised Fine-Tuning (SFT) is an effective method for adapting Large Language Models (LLMs) on downstream tasks. However, variability in training data can hinder a model's ability to generalize across domains. This paper studies the…

Computation and Language · Computer Science 2025-10-07 Davood Rafiei , Morgan Lindsay Heisler , Weiwei Zhang , Mohammadreza Pourreza , Yong Zhang

Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL) is a standard post-training recipe for improving Large Language Models (LLM) reasoning, but why it works remains unclear. We revisit the common claim that ``SFT memorizes,…

Machine Learning · Computer Science 2026-05-12 Hangzhan Jin , Sitao Luan , Tianwei Ni , Sicheng Lyu , Guillaume Rabusseau , Reihaneh Rabbany , Doina Precup , Mohammad Hamdaqa