中文
相关论文

相关论文: Eliminating Position Bias of Language Models: A Me…

200 篇论文

Large language models (LLMs) have demonstrated impressive capabilities in various tasks using the in-context learning (ICL) paradigm. However, their effectiveness is often compromised by inherent bias, leading to prompt brittleness, i.e.,…

计算与语言 · 计算机科学 2024-12-13 Hanzhang Zhou , Zijian Feng , Zixiao Zhu , Junlang Qian , Kezhi Mao

Fine-tuning has been demonstrated to be an effective method to improve the domain performance of large language models (LLMs). However, LLMs might fit the dataset bias and shortcuts for prediction, leading to poor generation performance.…

计算与语言 · 计算机科学 2024-07-02 Zhongkun Liu , Zheng Chen , Mengqi Zhang , Zhaochun Ren , Pengjie Ren , Zhumin Chen

Large language models (LLMs), even when specifically trained to process long input contexts, struggle to capture relevant information located in the middle of their input. This phenomenon has been known as the lost-in-the-middle problem. In…

Despite growing interest in using Large Language Models (LLMs) for educational assessment, it remains unclear how closely they align with human scoring. We present a systematic evaluation of instruction-tuned LLMs across three open…

计算与语言 · 计算机科学 2026-04-02 Filip J. Kucia , Anirban Chakraborty , Anna Wróblewska

Probing Pre-trained Language Models (PLMs) using prompts has indirectly implied that language models (LMs) can be treated as knowledge bases. To this end, this phenomena has been effective especially when these LMs are fine-tuned towards…

计算与语言 · 计算机科学 2022-04-08 M. Abaho , D. Bollegala , P. Williamson , S. Dodd

Positional bias in large language models (LLMs) hinders their ability to effectively process long inputs. A prominent example is the "lost in the middle" phenomenon, where LLMs struggle to utilize relevant information situated in the middle…

Large Language Models (LLMs) have demonstrated remarkable capabilities in executing tasks based on natural language queries. However, these models, trained on curated datasets, inherently embody biases ranging from racial to national and…

计算与语言 · 计算机科学 2024-07-29 Lynnette Hui Xian Ng , Iain Cruickshank , Roy Ka-Wei Lee

Large Language Models (LLMs) have demonstrated remarkable capabilities in various NLP tasks. However, previous works have shown these models are sensitive towards prompt wording, and few-shot demonstrations and their order, posing…

计算与语言 · 计算机科学 2023-08-23 Pouya Pezeshkpour , Estevam Hruschka

Positional bias - where models overemphasize certain positions regardless of content - has been shown to negatively impact model performance across various tasks. While recent research has extensively examined positional bias in text…

计算机视觉与模式识别 · 计算机科学 2025-11-17 Kebin Wu , Fatima Albreiki

LLM-as-a-Judge has emerged as a promising alternative to human evaluators across various tasks, yet inherent biases - particularly position bias, the tendency to favor solutions based on their position within the prompt - compromise its…

计算与语言 · 计算机科学 2025-11-12 Lin Shi , Chiyu Ma , Wenhua Liang , Xingjian Diao , Weicheng Ma , Soroush Vosoughi

Many extractive question answering models are trained to predict start and end positions of answers. The choice of predicting answers as positions is mainly due to its simplicity and effectiveness. In this study, we hypothesize that when…

计算与语言 · 计算机科学 2021-03-09 Miyoung Ko , Jinhyuk Lee , Hyunjae Kim , Gangwoo Kim , Jaewoo Kang

In this paper, we demonstrate that an inherent waveform pattern in the attention allocation of large language models (LLMs) significantly affects their performance in tasks demanding a high degree of context awareness, such as utilizing…

计算与语言 · 计算机科学 2024-06-05 Yuhan Chen , Ang Lv , Ting-En Lin , Changyu Chen , Yuchuan Wu , Fei Huang , Yongbin Li , Rui Yan

Alignment is a crucial step to enhance the instruction-following and conversational abilities of language models. Despite many recent work proposing new algorithms, datasets, and training pipelines, there is a lack of comprehensive studies…

计算与语言 · 计算机科学 2024-10-04 Xiao Yu , Qingyang Wu , Yu Li , Zhou Yu

Stance classification, the task of predicting the viewpoint of an author on a subject of interest, has long been a focal point of research in domains ranging from social science to machine learning. Current stance detection methods rely…

计算与语言 · 计算机科学 2024-03-07 Iain J. Cruickshank , Lynnette Hui Xian Ng

Language Models (LMs) assign significant attention to the first token, even if it is not semantically important, which is known as attention sink. This phenomenon has been widely adopted in applications such as streaming/long context…

计算与语言 · 计算机科学 2025-03-04 Xiangming Gu , Tianyu Pang , Chao Du , Qian Liu , Fengzhuo Zhang , Cunxiao Du , Ye Wang , Min Lin

The growing adoption of large language models (LLMs) in finance exposes high-stakes decision-making to subtle, underexamined positional biases. The complexity and opacity of modern model architectures compound this risk. We present the…

计算金融 · 定量金融 2025-10-08 Fabrizio Dimino , Krati Saxena , Bhaskarjit Sarmah , Stefano Pasquali

Large language models (LLMs) are capable of performing conditional sequence generation tasks, such as translation or summarization, through instruction fine-tuning. The fine-tuning data is generally sequentially concatenated from a specific…

计算与语言 · 计算机科学 2023-08-24 Yijin Liu , Xianfeng Zeng , Fandong Meng , Jie Zhou

The remarkable advancements in large language models (LLMs) have brought about significant improvements in Natural Language Processing(NLP) tasks. This paper presents a comprehensive review of in-context learning techniques, focusing on…

计算与语言 · 计算机科学 2023-09-26 Yinheng Li

Modern large language models (LLMs) are typically trained and deployed using structured role tags (e.g. system, user, assistant, tool) that explicitly mark the source of each piece of context. While these tags are essential for instruction…

计算与语言 · 计算机科学 2026-04-21 Xu Pan , Jingxuan Fan , Zidi Xiong , Ely Hahami , Jorin Overwiening , Ziqian Xie

Advancements in Large Language Models (LLMs) have increased the performance of different natural language understanding as well as generation tasks. Although LLMs have breached the state-of-the-art performance in various tasks, they often…