The recent developments in the research area of Large Language Models (LLMs) and their applications in software development and hardware design have shown significant advancements. The field is moving towards more adaptive and efficient solutions, leveraging LLMs for tasks such as code generation, refinement, and hardware trojan design. Innovations in carbon-aware computing strategies and geospatial sustainability assessments are also notable, reflecting a growing emphasis on environmental sustainability in technological advancements. Notably, the integration of LLMs with formal verification tools for generating formally verified code and the exploration of multi-agent collaboration in incident response are pushing the boundaries of what is possible in autonomous systems and cybersecurity. The field is also witnessing a shift towards more open-source and customizable solutions, addressing privacy concerns and reducing costs associated with proprietary APIs. The noteworthy papers in this area include 'VeCoGen: Automating Generation of Formally Verified C Code with Large Language Models,' which demonstrates the potential of combining LLMs with formal verification for automated program generation, and 'FaaSRCA: Full Lifecycle Root Cause Analysis for Serverless Applications,' which introduces a novel method for analyzing serverless applications across their entire lifecycle.
Advances in LLM Applications for Software and Hardware Development
Sources
Geospatial sustainability assessment of universal Fiber-To-The-Neighborhood (FTTnb) broadband infrastructure strategies for Sub-Saharan Africa
The Impact of Example Selection in Few-Shot Prompting on Automated Essay Scoring Using GPT Models
Quality Time: Carbon-Aware Quality Adaptation for Energy-Intensive Services
Using a Feedback Loop for LLM-based Infrastructure as Code Generation
VeCoGen: Automating Generation of Formally Verified C Code with Large Language Models
On the Adversarial Robustness of Instruction-Tuned Large Language Models for Code
Amplifying human performance in combinatorial competitive programming
Action Engine: An LLM-based Framework for Automatic FaaS Workflow Generation
o1-Coder: an o1 Replication for Coding
C2HLSC: Leveraging Large Language Models to Bridge the Software-to-Hardware Design Gap
Analyzing the Energy and Accuracy of LLMs in Software Development
Towards Adaptive Mechanism Activation in Language Agent
Multi-Agent Collaboration in Incident Response with Large Language Models
Free and Customizable Code Documentation with LLMs: A Fine-Tuning Approach
ChatCollab: Exploring Collaboration Between Humans and AI Agents in Software Teams
FaaSRCA: Full Lifecycle Root Cause Analysis for Serverless Applications
Exploring the Potential of Llama Models in Automated Code Refinement: A Replication Study
Unleashing GHOST: An LLM-Powered Framework for Automated Hardware Trojan Design
GoldFish: Serverless Actors with Short-Term Memory State for the Edge-Cloud Continuum
Does Few-Shot Learning Help LLM Performance in Code Synthesis?
Revolve: Optimizing AI Systems by Tracking Response Evolution in Textual Optimization
Controlling the Mutation in Large Language Models for the Efficient Evolution of Algorithms
PerfCodeGen: Improving Performance of LLM Generated Code with Execution Feedback
How to Correctly do Semantic Backpropagation on Language-based Agentic Systems
A Water Efficiency Dataset for African Data Centers
From Code to Play: Benchmarking Program Search for Games Using Large Language Models