中文
相关论文

相关论文: HoneyBee: Progressive Instruction Finetuning of La…

200 篇论文

With the rapid scaling of large language models (LLMs), structured pruning has become a widely used technique to learn efficient, smaller models from larger ones, delivering superior performance compared to training similarly sized models…

计算与语言 · 计算机科学 2025-06-04 Bairu Hou , Qibin Chen , Jianyu Wang , Guoli Yin , Chong Wang , Nan Du , Ruoming Pang , Shiyu Chang , Tao Lei

High-quality long-context instruction data is essential for aligning long-context large language models (LLMs). Despite the public release of models like Qwen and Llama, their long-context instruction data remains proprietary. Human…

计算与语言 · 计算机科学 2025-06-04 Chaochen Gao , Xing Wu , Zijia Lin , Debing Zhang , Songlin Hu

Training large language models (LLMs) to follow instructions has significantly enhanced their ability to tackle unseen tasks. However, despite their strong generalization capabilities, instruction-following LLMs encounter difficulties when…

计算与语言 · 计算机科学 2025-05-29 Maja Stahl , Timon Ziegenbein , Joonsuk Park , Henning Wachsmuth

Large vision-language models (LVLMs) have shown promising performance on a variety of vision-language tasks. However, they remain susceptible to hallucinations, generating outputs misaligned with visual content or instructions. While…

计算机视觉与模式识别 · 计算机科学 2024-07-17 Jinrui Zhang , Teng Wang , Haigang Zhang , Ping Lu , Feng Zheng

The rapid evolution of artificial intelligence in drug discovery encounters challenges with generalization and extensive training, yet Large Language Models (LLMs) offer promise in reshaping interactions with complex molecular data. Our…

生物大分子 · 定量生物学 2024-12-20 He Cao , Zijing Liu , Xingyu Lu , Yuan Yao , Yu Li

It is crucial for large language models (LLMs) to follow instructions that involve multiple constraints. However, it is an unexplored area to enhance LLMs' ability to follow soft constraints. To bridge the gap, we initially design a…

计算与语言 · 计算机科学 2025-06-03 Qingyu Ren , Jie Zeng , Qianyu He , Jiaqing Liang , Yanghua Xiao , Weikang Zhou , Zeye Sun , Fei Yu

There is a consensus that instruction fine-tuning of LLMs requires high-quality data, but what are they? LIMA (NeurIPS 2023) and AlpaGasus (ICLR 2024) are state-of-the-art methods for selecting such high-quality examples, either via manual…

计算与语言 · 计算机科学 2024-06-05 Hao Zhao , Maksym Andriushchenko , Francesco Croce , Nicolas Flammarion

We present the Open MatSci ML Toolkit: a flexible, self-contained, and scalable Python-based framework to apply deep learning models and methods on scientific data with a specific focus on materials science and the OpenCatalyst Dataset. Our…

机器学习 · 计算机科学 2023-09-01 Santiago Miret , Kin Long Kelvin Lee , Carmelo Gonzales , Marcel Nassar , Matthew Spellings

Large language models respond well in high-resource languages like English but struggle in low-resource languages. It may arise from the lack of high-quality instruction following data in these languages. Directly translating English…

计算与语言 · 计算机科学 2024-05-31 Chong Li , Wen Yang , Jiajun Zhang , Jinliang Lu , Shaonan Wang , Chengqing Zong

Large language models have achieved significant success in various domains, yet their understanding of lyric-centric knowledge has not been fully explored. In this work, we first introduce PlaylistSense, a dataset to evaluate the playlist…

计算与语言 · 计算机科学 2026-01-06 Jiani Guo , Jiajia Li , Jie Wu , Zuchao Li , Yujiu Yang , Ping Wang

Large language models excel at instruction-following in English, but their performance in low-resource languages like Thai remains underexplored. Existing benchmarks often rely on translations, missing cultural and domain-specific nuances…

Large language models (LLMs) often produce inaccurate or misleading content-hallucinations. To address this challenge, we introduce Noise-Augmented Fine-Tuning (NoiseFiT), a novel framework that leverages adaptive noise injection based on…

计算与语言 · 计算机科学 2025-05-06 Afshin Khadangi , Amir Sartipi , Igor Tchappi , Ramin Bahmani

Language models often pre-train on large unsupervised text corpora, then fine-tune on additional task-specific data. However, typical fine-tuning schemes do not prioritize the examples that they tune on. We show that, if you can prioritize…

计算与语言 · 计算机科学 2023-05-12 Ian Osband , Seyed Mohammad Asghari , Benjamin Van Roy , Nat McAleese , John Aslanides , Geoffrey Irving

Training automatic summary fact verifiers often faces the challenge of a lack of human-labeled data. In this paper, we explore alternative way of leveraging Large Language Model (LLM) generated feedback to address the inherent limitation of…

计算与语言 · 计算机科学 2024-12-17 Jihwan Oh , Jeonghwan Choi , Nicole Hee-Yeon Kim , Taewon Yun , Hwanjun Song

Solving open-ended science questions remains challenging for large language models, particularly due to inherently unreliable supervision and evaluation. The bottleneck lies in the data construction and reward design for scientific…

计算与语言 · 计算机科学 2026-02-11 Zijie Chen , Zhenghao Lin , Xiao Liu , Zhenzhong Lan , Yeyun Gong , Peng Cheng

As Large Language Models (LLMs) are rapidly evolving, providing accurate feedback and scalable oversight on their outputs becomes an urgent and critical problem. Leveraging LLMs as critique models to achieve automated supervision is a…

计算与语言 · 计算机科学 2025-05-02 Wenkai Yang , Jingwen Chen , Yankai Lin , Ji-Rong Wen

Mathematical reasoning has been challenging for large language models (LLMs), and the introduction of step-by-step Chain-of-Thought (CoT) inference has significantly advanced the mathematical capabilities of LLMs. However, current…

人工智能 · 计算机科学 2025-09-23 Lang Cao , Yingtian Zou , Chao Peng , Renhong Chen , Wu Ning , Yitong Li

Materials discovery and development are critical for addressing global challenges. Yet, the exponential growth in materials science literature comprising vast amounts of textual data has created significant bottlenecks in knowledge…

Multimodal large language models (MLLMs) have demonstrated remarkable capabilities in vision-language answering tasks. Despite their strengths, these models often encounter challenges in achieving complex reasoning tasks such as…

人工智能 · 计算机科学 2025-11-11 Jinhao Chen , Zhen Yang , Jianxin Shi , Tianyu Wo , Jie Tang

Modern large language models often encode sensitive, harmful, or copyrighted knowledge, raising the need for post-hoc unlearning-the ability to remove specific domains of knowledge from a model without full retraining. A major bottleneck in…

计算与语言 · 计算机科学 2025-10-08 Xiaoyuan Zhu , Muru Zhang , Ollie Liu , Robin Jia , Willie Neiswanger