We gratefully acknowledge support from
the Simons Foundation and member institutions.

Computation and Language

Authors and titles for recent submissions

[ total of 343 entries: 1-25 | 26-50 | 51-75 | 76-100 | ... | 326-343 ]
[ showing 25 entries per page: fewer | more | all ]

Thu, 25 Apr 2024 (showing first 25 of 47 entries)

[1]  arXiv:2404.16020 [pdf, other]
Title: Universal Adversarial Triggers Are Not Universal
Subjects: Computation and Language (cs.CL)
[2]  arXiv:2404.16019 [pdf, other]
Title: The PRISM Alignment Project: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models
Subjects: Computation and Language (cs.CL)
[3]  arXiv:2404.15949 [pdf, other]
Title: Sequence can Secretly Tell You What to Discard
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
[4]  arXiv:2404.15928 [pdf, other]
Title: Generalization Measures for Zero-Shot Cross-Lingual Transfer
Subjects: Computation and Language (cs.CL)
[5]  arXiv:2404.15894 [pdf, other]
Title: Assessing The Potential Of Mid-Sized Language Models For Clinical QA
Comments: 25 pages, 8 figures
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
[6]  arXiv:2404.15877 [pdf, other]
Title: Effective Unsupervised Constrained Text Generation based on Perturbed Masking
Subjects: Computation and Language (cs.CL)
[7]  arXiv:2404.15848 [pdf, other]
Title: Detecting Conceptual Abstraction in LLMs
Comments: Paper accepted at the LREC-COLING 2024 Conference (Paper ID: 1968) this https URL
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
[8]  arXiv:2404.15846 [pdf, other]
Title: From Complex to Simple: Enhancing Multi-Constraint Complex Instruction Following Ability of Large Language Models
Subjects: Computation and Language (cs.CL)
[9]  arXiv:2404.15845 [pdf, other]
Title: Exploring LLM Prompting Strategies for Joint Essay Scoring and Feedback Generation
Comments: Accepted to BEA Workshop 2024
Subjects: Computation and Language (cs.CL)
[10]  arXiv:2404.15807 [pdf, other]
Title: One Subgraph for All: Efficient Reasoning on Opening Subgraphs for Inductive Knowledge Graph Completion
Subjects: Computation and Language (cs.CL)
[11]  arXiv:2404.15777 [pdf, ps, other]
Title: A Comprehensive Survey on Evaluating Large Language Model Applications in the Medical Industry
Comments: 18 pages
Subjects: Computation and Language (cs.CL)
[12]  arXiv:2404.15758 [pdf, other]
Title: Let's Think Dot by Dot: Hidden Computation in Transformer Language Models
Comments: 17 pages, 10 figures
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
[13]  arXiv:2404.15737 [pdf, other]
Title: No Train but Gain: Language Arithmetic for training-free Language Adapters enhancement
Subjects: Computation and Language (cs.CL)
[14]  arXiv:2404.15720 [pdf, other]
Title: Annotator-Centric Active Learning for Subjective NLP Tasks
Subjects: Computation and Language (cs.CL)
[15]  arXiv:2404.15702 [pdf, other]
Title: Nyonic Technical Report
Subjects: Computation and Language (cs.CL)
[16]  arXiv:2404.15690 [pdf, other]
Title: Neural Proto-Language Reconstruction
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
[17]  arXiv:2404.15676 [pdf, other]
Title: Beyond Chain-of-Thought: A Survey of Chain-of-X Paradigms for LLMs
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
[18]  arXiv:2404.15667 [pdf, other]
Title: The Promise and Challenges of Using LLMs to Accelerate the Screening Process of Systematic Reviews
Comments: Accepted to the International Conference on Evaluation and Assessment in Software Engineering (EASE), 2024 edition
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
[19]  arXiv:2404.15660 [pdf, other]
Title: KS-LLM: Knowledge Selection of Large Language Models with Evidence Document for Question Answering
Subjects: Computation and Language (cs.CL)
[20]  arXiv:2404.15650 [pdf, other]
Title: Return of EM: Entity-driven Answer Set Expansion for QA Evaluation
Comments: Under Review (9 pages, 3 figures)
Subjects: Computation and Language (cs.CL)
[21]  arXiv:2404.15639 [pdf, other]
Title: CodeIP: A Grammar-Guided Multi-Bit Watermark for Large Language Models of Code
Comments: 13 pages, 7 figures
Subjects: Computation and Language (cs.CL)
[22]  arXiv:2404.15604 [pdf, ps, other]
Title: Hybrid LLM/Rule-based Approaches to Business Insights Generation from Structured Data
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
[23]  arXiv:2404.15588 [pdf, other]
Title: Minimal Evidence Group Identification for Claim Verification
Subjects: Computation and Language (cs.CL)
[24]  arXiv:2404.15578 [pdf, ps, other]
Title: Can Foundational Large Language Models Assist with Conducting Pharmaceuticals Manufacturing Investigations?
Authors: Hossein Salami (1), Brandye Smith-Goettler (2), Vijay Yadav (2) ((1) Digital Services, MMD, Merck & Co., Inc., Rahway, NJ, USA, (2) Digital Services, MMD, Merck & Co., Inc., West Point, PA, USA)
Comments: 13 pages, 3 figures
Subjects: Computation and Language (cs.CL)
[25]  arXiv:2404.15574 [pdf, other]
Title: Retrieval Head Mechanistically Explains Long-Context Factuality
Comments: Preprint
Subjects: Computation and Language (cs.CL)
[ total of 343 entries: 1-25 | 26-50 | 51-75 | 76-100 | ... | 326-343 ]
[ showing 25 entries per page: fewer | more | all ]

Disable MathJax (What is MathJax?)

Links to: arXiv, form interface, find, cs, new, 2404, contact, help  (Access key information)