中文
相关论文

相关论文: Lost in Sampling: Assessing Lexical Reachability i…

200 篇论文

Large Language Models (LLMs) are widely deployed in real-world applications, yet little is known about their training dynamics at the token level. Evaluation typically relies on aggregated training loss, measured at the batch level, which…

计算与语言 · 计算机科学 2024-10-17 Andrea Pinto , Tomer Galanti , Randall Balestriero

Large Language Models (LLMs) are emerging as transformative tools for software vulnerability detection, addressing critical challenges in the security domain. Traditional methods, such as static and dynamic analysis, often falter due to…

密码学与安全 · 计算机科学 2025-02-19 Ze Sheng , Zhicheng Chen , Shuning Gu , Heqing Huang , Guofei Gu , Jeff Huang

A common practice in large language model (LLM) usage for complex analytical tasks such as code generation, is to sample a solution for the entire task within the model's context window. Previous works have shown that subtask decomposition…

人工智能 · 计算机科学 2025-02-03 Yotam Wolf , Binyamin Rothberg , Dorin Shteyman , Amnon Shashua

The performance of conversational Large Language Models (LLMs) in general, and of ChatGPT in particular, is currently being evaluated on many different tasks, from logical reasoning or maths to answering questions on a myriad of topics.…

计算与语言 · 计算机科学 2024-10-22 Gonzalo Martínez , José Alberto Hernández , Javier Conde , Pedro Reviriego , Elena Merino

Large Language Models (LLMs) have emerged as a promising cornerstone for the development of natural language processing (NLP) and artificial intelligence (AI). However, ensuring the robustness of LLMs remains a critical challenge. To…

计算与语言 · 计算机科学 2025-11-07 Pankaj Kumar , Subhankar Mishra

Large language models (LLMs) are deployed in a wide variety of user-facing applications. Typically, these deployments have some specific purpose, like answering questions grounded on documentation or acting as coding assistants, but they…

计算与语言 · 计算机科学 2025-11-14 David Yunis , Siyu Huo , Chulaka Gunasekara , Danish Contractor

Any piece of knowledge is usually expressed in one or a handful of natural languages on the web or in any large corpus. Large Language Models (LLMs) act as a bridge by acquiring knowledge from a source language and making it accessible when…

计算与语言 · 计算机科学 2025-10-20 Vihari Piratla , Purvam Jain , Darshan Singh , Partha Talukdar , Trevor Cohn

The widespread adoption of large language models (LLMs) necessitates reliable methods to detect LLM-generated text. We introduce SimMark, a robust sentence-level watermarking algorithm that makes LLMs' outputs traceable without requiring…

计算与语言 · 计算机科学 2025-09-12 Amirhossein Dabiriaghdam , Lele Wang

Large language models are increasingly deployed in settings where reliability matters, yet output-level uncertainty signals such as token probabilities, entropy, and self-consistency can become brittle under calibration--deployment…

计算与语言 · 计算机科学 2026-04-20 Yanli Wang , Peng Kuang , Xiaoyu Han , Kaidi Xu , Haohan Wang

This study investigates semantic uncertainty in large language model (LLM) outputs across different decoding methods, focusing on emerging techniques like speculative sampling and chain-of-thought (CoT) decoding. Through experiments on…

计算与语言 · 计算机科学 2025-06-24 Darius Foodeei , Simin Fan , Martin Jaggi

For Pretrained Language Models (PLMs), their susceptibility to noise has recently been linked to subword segmentation. However, it is unclear which aspects of segmentation affect their understanding. This study assesses the robustness of…

计算与语言 · 计算机科学 2024-10-15 Xinzhe Li , Ming Liu , Shang Gao

Large language models (LLMs) demonstrate exceptional performance on tasks requiring complex linguistic abilities, such as reference disambiguation and metaphor recognition/generation. Although LLMs possess impressive capabilities, their…

计算与语言 · 计算机科学 2025-09-16 Yi Jing , Zijun Yao , Hongzhu Guo , Lingxu Ran , Xiaozhi Wang , Lei Hou , Juanzi Li

This chapter explores advancements in decoding strategies for large language models (LLMs), focusing on enhancing the Locally Typical Sampling (LTS) algorithm. Traditional decoding methods, such as top-k and nucleus sampling, often struggle…

计算与语言 · 计算机科学 2025-06-12 Jaydip Sen , Saptarshi Sengupta , Subhasis Dasgupta

Context: Large language models (LLMs) are increasingly used to screen literature for systematic reviews (SRs), but the standard confusion-matrix metrics used to evaluate them can mislead under the imbalanced, cost-asymmetric conditions of…

软件工程 · 计算机科学 2026-04-28 Lech Madeyski , Barbara Kitchenham , Martin Shepperd

Large Language Models (LLMs) effectiveness is usually evaluated by means of benchmarks such as MMLU, ARC-C, or HellaSwag, where questions are presented in their original wording, thus in a fixed, standardized format. However, real-world…

计算与语言 · 计算机科学 2025-09-05 Riccardo Lunardi , Vincenzo Della Mea , Stefano Mizzaro , Kevin Roitero

Large Language Models (LLMs) achieve strong linguistic performance, yet their internal mechanisms for producing these predictions remain unclear. We investigate the hypothesis that LLMs encode representations of linguistic constraint…

计算与语言 · 计算机科学 2026-05-15 Hardy , Sebastian Padó

Multilingual Large Language Models (LLMs) exhibit remarkable cross-lingual abilities, yet often exhibit a systematic bias toward the representations from other languages, resulting in semantic interference when generating content in…

计算与语言 · 计算机科学 2026-01-21 Ilia Badanin , Daniil Dzenhaliou , Imanol Schlag

The advent of large language models (LLMs) has dramatically advanced the state-of-the-art in numerous natural language generation tasks. For LLMs to be applied reliably, it is essential to have an accurate measure of their confidence.…

计算与语言 · 计算机科学 2024-06-05 Zhen Lin , Shubhendu Trivedi , Jimeng Sun

The capabilities of large language models have grown significantly in recent years and so too have concerns about their misuse. It is important to be able to distinguish machine-generated text from human-authored content. Prior works have…

密码学与安全 · 计算机科学 2024-10-15 Julien Piet , Chawin Sitawarin , Vivian Fang , Norman Mu , David Wagner

As Large Language Models (LLMs) become increasingly prevalent, their generated outputs are proliferating across the web, risking a future where machine-generated content dilutes human-authored text. Since online data is the primary resource…

计算与语言 · 计算机科学 2025-09-23 George Drayson , Emine Yilmaz , Vasileios Lampos