The Science of Detecting LLM-Generated Text
A technical introduction to LLM-generated text detection methods.
Research and Advances
Advertisement
Indistinguishability Obfuscation from Well-Founded Assumptions
We examine a formalization of the “one-way compiler" concept with the notion of indistinguishability obfuscation.
Co-Developing Programs and Their Proof of Correctness
There have been some successes with program verification systems, but they are still perceived as a niche activity for the most critical and specialized projects.
A Unifying Framework for Incompleteness, Inconsistency, and Uncertainty in Databases
Deploying possible world semantics and the challenge of computing the certain answers to queries.
Measuring GitHub Copilot’s Impact on Productivity
A case study asks Copilot users about the tool's impact on their productivity, and seeks to find their perceptions mirrored in user data.
Taming Algorithmic Priority Inversion in Mission-Critical Perception Pipelines
This paper discusses algorithmic priority inversion in mission-critical machine inference pipelines used in modern neural-network-based perception subsystems and describes a solution to mitigate its effect.
Technical Perspective: Bridging AI with Real-Time Systems
"Taming Algorithmic Priority Inversion in Mission-Critical Perception Pipelines," by Shengzhong Liu et al., proposes a new methodology for overcoming the limitations of current AI frameworks to enable the use of deep neural networks in mission-critical systems.
Superpolynomial Lower Bounds Against Low-Depth Algebraic Circuits
In this paper, we prove the first superpolynomial lower bounds against algebraic circuits of all constant depths over all fields of characteristic 0.
Technical Perspective: How Easy Is It to Describe Hard Polynomials?
"Superpolynomial Lower Bounds Against Low-Depth Algebraic Circuits," by Nutan Limaye et al., achieves a landmark in the larger quest of understanding hardness, dentity testing, and reconstruction.
Anthropomorphism and Human-Robot Interaction
Humans are more likely to interact with robots if the robots look human.
Talking about Large Language Models
It is a serious mistake to apply to AI systems the same intuitions that we deploy in our dealings with each other.
Energy and Emissions of Machine Learning on Smartphones vs. the Cloud
An extensive study from Google finds machine-learning training in the cloud can significantly reduce carbon footprints.
Computing Education in the Era of Generative AI
Computing educators and students face challenges and opportunities in adapting to LLMs capable of generating accurate source code from natural-language problem descriptions.
Inherent Limitations of AI Fairness
AI fairness should not be considered a panacea: It may have the potential to make society more fair than ever, but it needs critical thought and outside help to make it happen.
Nonverbal Communication through Expressive Objects
In this work, we work closely with an AAC user to understand how motion through a physical expressive object can support their communication.
The User as a Key Ingredient in AAC Design
"Nonverbal Communication through Expressive Objects," by Stephanie Valencia et al., describes the design of a physical object to support communication and focuses on the active involvement of an expert Augmentative and Alternative Communication (AAC) user as a member of the design team.
Shortcut Learning of Large Language Models in Natural Language Understanding
Shortcuts often hinder the robustness of large language models.
On Specifying for Trustworthiness
As autonomous systems increasingly become part of our lives, it is crucial to foster trust between humans and these systems, to ensure positive outcomes and mitigate harmful ones.
What Should We Do when Our Ideas of Fairness Conflict?
Standards for fair decision making could help us develop algorithms that comport with our consensus views; however, algorithmic fairness has its limits.
10 Things Software Developers Should Learn about Learning
Understanding how human memory and learning works, the differences between beginners and experts, and practical steps developers can take to improve their learning, training, and recruitment.
Understanding and handling interference across multiple active cameras.
Theoretical Analysis of Edit Distance Algorithms
To what extent have the techniques for theoretical analysis of edit distance algorithms achieved their goals?
Data Analytics Anywhere and Everywhere
Mobile, ubiquitous, and immersive computing appear poised to transform visualization, data science, and data-driven decision making.
Shape the Future of Computing
ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.
Get InvolvedCommunications of the ACM (CACM) is now a fully Open Access publication.
By opening CACM to the world, we hope to increase engagement among the broader computer science community and encourage non-members to discover the rich resources ACM has to offer.
Learn More