English
Related papers

Related papers: Paraphrase-Induced Output-Mode Collapse: When LLMs…

200 papers

Guard models are a critical component of LLM safety, but their sensitivity to superficial linguistic variations remains a key vulnerability. We show that even meaning-preserving paraphrases can cause large fluctuations in safety scores,…

Computation and Language · Computer Science 2025-11-17 Cristina Pinneri , Christos Louizos

Due to their architecture and vast pre-training data, large language models (LLMs) demonstrate strong text classification performance. However, LLM output - here, the category assigned to a text - depends heavily on the wording of the…

Computation and Language · Computer Science 2025-12-04 Kylie L. Anglin , Stephanie Milan , Brittney Hernandez , Claudia Ventura

The drastic increase of large language models' (LLMs) parameters has led to a new research direction of fine-tuning-free downstream customization by prompts, i.e., task descriptions. While these prompt-based services (e.g. OpenAI's GPTs)…

Computation and Language · Computer Science 2025-02-13 Zi Liang , Haibo Hu , Qingqing Ye , Yaxin Xiao , Haoyang Li

Intrinsic self-correct was a method that instructed large language models (LLMs) to verify and correct their responses without external feedback. Unfortunately, the study concluded that the LLMs could not self-correct reasoning yet. We find…

Computation and Language · Computer Science 2024-10-04 Zhenyu Wu , Qingkai Zeng , Zhihan Zhang , Zhaoxuan Tan , Chao Shen , Meng Jiang

Large Language Models (LLMs) have demonstrated substantial progress in biomedical and clinical applications, motivating rigorous evaluation of their ability to answer nuanced, evidence-based questions. We curate a multi-source benchmark…

Computation and Language · Computer Science 2025-09-16 Can Wang , Yiqun Chen

Large Language Models (LLMs) are regularly being used to label data across many domains and for myriad tasks. By simply asking the LLM for an answer, or ``prompting,'' practitioners are able to use LLMs to quickly get a response for an…

Computation and Language · Computer Science 2024-04-03 Abel Salinas , Fred Morstatter

Large Language Models (LLMs) currently respond to every prompt. However, they can produce incorrect answers when they lack knowledge or capability -- a problem known as hallucination. We instead propose post-training an LLM to generate…

Computation and Language · Computer Science 2026-02-17 Tim Franzmeyer , Archie Sravankumar , Lijuan Liu , Yuning Mao , Rui Hou , Sinong Wang , Jakob N. Foerster , Luke Zettlemoyer , Madian Khabsa

Benchmarks for large language models (LLMs) often rely on rubric-scented prompts that request visible reasoning and strict formatting, whereas real deployments demand terse, contract-bound answers. We investigate whether such "evaluation…

Computation and Language · Computer Science 2025-10-13 Nisar Ahmed , Muhammad Imran Zaman , Gulshan Saleem , Ali Hassan

Recent work reports that Large Reasoning Models (LRMs) undergo a collapse in performance on solving puzzles beyond certain perplexity thresholds. In subsequent discourse, questions have arisen as to whether the nature of the task muddles an…

Artificial Intelligence · Computer Science 2025-10-21 Chris Su , Harrison Li , Matheus Marques , George Flint , Kevin Zhu , Sunishchal Dev

Medical reasoning models (MRMs) achieve superior performance on medical benchmarks compared to medical LLMs; however, high accuracy alone is insufficient for practical deployment. One of such requirements for real-world application is…

Large Language Model (LLM) agents are increasingly expected to maintain coherent, long-term personalized memory, yet current benchmarks primarily measure static fact retrieval, overlooking the ability to revise stored beliefs when new…

Computation and Language · Computer Science 2026-05-08 Hanxiang Chao , Yihan Bai , Rui Sheng , Tianle Li , Yushi Sun

Large language models (LLMs) produce context inconsistency hallucinations, which are LLM generated outputs that are misaligned with the user prompt. This research project investigates whether prompt engineering (PE) methods can mitigate…

Computation and Language · Computer Science 2025-12-19 Imane Jaaouine , Ross D. King

An interesting behavior in large language models (LLMs) is prompt sensitivity. When provided with different but semantically equivalent versions of the same prompt, models may produce very different distributions of answers. This suggests…

Computation and Language · Computer Science 2025-10-21 Kyle Cox , Jiawei Xu , Yikun Han , Rong Xu , Tianhao Li , Chi-Yang Hsu , Tianlong Chen , Walter Gerych , Ying Ding

Large Language Models (LLMs) are increasingly deployed for open-domain question answering, yet their alignment with human perspectives on temporally recent information remains underexplored. We introduce RECOM (Reddit Evaluation for…

Computation and Language · Computer Science 2026-02-13 Pushwitha Krishnappa , Amit Das , Vinija Jain , Tathagata Mukherjee , Aman Chadha

Large Language Models (LLMs) are increasingly vulnerable to a sophisticated form of adversarial prompting known as camouflaged jailbreaking. This method embeds malicious intent within seemingly benign language to evade existing safety…

Cryptography and Security · Computer Science 2025-09-09 Youjia Zheng , Mohammad Zandsalimy , Shanu Sushmita

Large Language Models (LLMs) demonstrate strong mathematical problem-solving abilities but frequently fail on problems that deviate syntactically from their training distribution. We identify a systematic failure mode, syntactic blind…

Computation and Language · Computer Science 2025-10-03 Dane Williamson , Yangfeng Ji , Matthew Dwyer

Despite significant research effort in the development of automatic dialogue evaluation metrics, little thought is given to evaluating dialogues other than in English. At the same time, ensuring metrics are invariant to semantically similar…

Computation and Language · Computer Science 2023-09-11 John Mendonça , Patrícia Pereira , Helena Moniz , João Paulo Carvalho , Alon Lavie , Isabel Trancoso

We evaluate large language models (LLMs) for automatic personality prediction from text under the binary Five Factor Model (BIG5). Five models -- including GPT-4 and lightweight open-source alternatives -- are tested across three…

Computation and Language · Computer Science 2025-12-01 Francesco Di Cursi , Chiara Boldrini , Marco Conti , Andrea Passarella

With the increasing use of large language models (LLMs), ensuring reliable performance in diverse, real-world environments is essential. Despite their remarkable achievements, LLMs often struggle with adversarial inputs, significantly…

Computation and Language · Computer Science 2024-06-18 Yuqing Wang , Yun Zhao

Large language models (LLMs) achieve promising results in code generation based on a given natural language description. They have been integrated into open-source projects and commercial products to facilitate daily coding activities. The…

Software Engineering · Computer Science 2024-07-01 Junkai Chen , Zhenhao Li , Xing Hu , Xin Xia