#prompt engineering

try —

45 papers match

cs.AI2026

What makes prompts a graph: necessary and sufficient conditions for prompt graph engineering

Sandeco Macedo

The paper defines prompts as nodes in an explicit, executable graph and establishes four necessary conditions for prompt graph engineering, providing a formal definition and vocabu…

#prompt engineering#graph structures#language model orchestration#AI system design
cs.CL2026

Challenges in annotations by humans and LLMs: A case study of evaluative language

Mirela Imamovic, Aenne Cecilia Kristine Knierim, Khushi Pitroda +1

The paper compares human annotators (trained linguist and trainees) with large language models on labeling evaluative language in TED talk transcripts using Appraisal theory, devel…

#annotation#large language models#evaluative language#appraisal theory
cs.SE2026

From Textual Requirements to Microservice Architectures - A Comprehensive Evaluation of LLM-Based Design Synthesis

Danyllo Albuquerque, José Renan, Guillermo Rodríguez +5

The paper evaluates whether a large language model (OpenAI o3) can automatically generate microservice architectures from textual requirements, comparing zero-shot and few-shot pro…

#microservice architecture#requirements engineering#large language models#prompt engineering
cs.CL2026

Would You Walk to the Car Wash? Revealing the Salience Bias of Large Language Models in Commonsense Reasoning

Zheng Wu, Chenhao Xue, Shijie Zheng +3

The paper identifies a "salience bias" in large language models where explicit but irrelevant details cause the models to overlook implicit commonsense knowledge, and shows that th…

#commonsense reasoning#large language models#salience bias#prompt engineering
cs.DL2026

Scientific Knowledge Discovery in the Age of Large Language Models

Eleni Adamidi, Serafeim Chatzopoulos, Thanasis Vergoulis

The paper surveys 34 peer‑reviewed studies that apply generative large language models to automate scientific literature retrieval and eligibility screening, analyzing model choice…

#literature search#eligibility screening#large language models#systematic review
cs.LG2026

Two Calls Beat Five Agents: Evaluating Multi-Agent Pipelines Against Self-Refinement for Local Language Models

Ashish Prajapati, Om Mohite

The paper compares a five‑role multi‑agent LLM pipeline with a simpler two‑call self‑refinement approach on a local 7B model, finding that communication format and implementation d…

#multi-agent systems#self‑refinement#local language models#prompt engineering