185 articles tagged with #prompt-engineering. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.
Researchers propose a debiasing fine-tuning method to improve Large Language Model robustness against semantically-neutral prompt variations without expensive full retraining. The approach identifies perturbation-induced bias in neural network outputs and demonstrates theoretical and experimental evidence that targeted debiasing can enhance model resilience to prompt alterations.
Researchers present methods for improving how large language models generate diverse pools of creative ideas during parallel inference without relying on seed examples. Their findings show that semantic direction stratification—organizing generations across different semantic directions with a single planning call—outperforms anchor-dependent baselines while maintaining quality and computational efficiency.
A longitudinal study examined how AI models (Gemini and Coteach) perform on mathematics task classification using the Task Analysis Guide, testing stability across model versions and responsiveness to few-shot prompting. Results showed newer model versions produced mixed effects, but few-shot prompting consistently improved both models' accuracy, suggesting prompt engineering is more reliable than passive model updates for specialized educational tasks.
Researchers propose a novel method for optimizing multi-agent LLM systems by decomposing credit assignment into temporal and structural components, enabling more efficient prompt optimization through targeted refinement rather than global updates. The approach uses state-space bottleneck analysis and role-based policy isolation to identify and fix weak components in collaborative AI systems, reducing computational queries while improving reasoning performance across benchmarks.
Researchers propose Canonical-Context On-Policy Distillation (CCOPD), a training method that improves large language models' ability to solve problems when information is revealed incrementally across multiple conversation turns rather than all at once. By using a frozen teacher model with complete context to guide a student model receiving fragmented information, CCOPD achieves 32% relative performance improvement on multi-turn tasks while maintaining single-prompt performance.
Researchers introduce SWAI, a training-free method for controlling language model outputs by manipulating logit scores using corpus-derived statistics. The technique enables real-time steering of model behavior—such as adjusting readability, politeness, and toxicity—without modifying model weights or accessing internal layers, outperforming existing prompt-based and logit-level baselines.
Researchers propose a hierarchical framework for deploying compact language models in resource-constrained agentic systems, combining knowledge distillation with oracle-supervised fine-tuning to maintain protocol compliance and semantic performance. The approach addresses core deployment challenges including context length limitations, memory constraints, and cost efficiency by separating schema learning from semantic adaptation.
Researchers demonstrate that standard statistical hypothesis tests fail when applied to generative surveying, where LLM-based personas provide market research feedback. The study proposes a valid permutation test that accounts for prompt sensitivity and provides guidance on optimal resource allocation for this emerging research methodology.
Researchers benchmarked 43 large language models used for academic scholar recommendations, revealing that prompt design significantly affects recommendation quality and diversity. The study found that model choice, persona prompting (language, location, role), and context variables independently shape which scholars are recommended, with geographic location prompts producing the most variation in factuality and representativeness across disciplines.
Researchers present twelve token optimization strategies for using LLMs to migrate Oracle databases to PostgreSQL, addressing cost and quality degradation challenges. Adaptive routing emerges as the optimal approach, reducing token consumption by 8.72% while maintaining 88.40% semantic match accuracy, demonstrating that token optimization requires balancing multiple objectives rather than simple prompt shortening.
Researchers investigated why chain-of-thought prompting improves language model accuracy by analyzing what happens at inference time rather than generation time. They discovered that the improvement comes primarily from lexical activation and short-range token co-occurrence (2-3 adjacent tokens) rather than from logical sentence-level reasoning, challenging assumptions about how rationales actually drive model performance.
Researchers introduce Augment Engineering, a methodology for orchestrating multiple AI tools across professional domains by applying portable meta-skills like prompt and context engineering. A five-month case study demonstrates that a single practitioner can produce work traditionally requiring domain specialists across seven domains, with statistical evidence supporting increased efficiency and production acceleration.
Researchers reveal that correct demonstrations in in-context learning don't guarantee improved model performance—some accurate examples actually degrade accuracy. The study introduces task-preserving perturbations to show that exemplar utility depends on how demonstrations influence contextual inference, not merely on correctness, challenging conventional assumptions about how AI models learn from examples.
Researchers evaluated 13 large language models' ability to generate code following the Singleton design pattern across four prompting strategies, finding that iterative binary feedback and instruction-based guidance most effectively guide LLMs to incorporate architectural best practices while maintaining code functionality.
Researchers adapted Microsoft's QuantumKatas quantum computing curriculum from Q# to Qiskit and created a 350-task benchmark with LLM evaluation infrastructure. Testing 16 language models revealed significant capability gaps, with frontier models achieving 83.1% pass rates versus 32.3% for weaker models, while highlighting that LLMs excel at implementing known algorithms but struggle with problem encoding.
Researchers have developed a mechanistic interpretability framework that reverses information flow through Chain-of-Thought prompting to understand how AI models reason. The study reveals CoT functions as a decoding space pruner that uses answer templates to guide outputs, with task-dependent neuron modulation that reduces activation in open-domain tasks but increases it in closed-domain scenarios.
Researchers propose Adaptive Multi-prompt Contrastive Network (AMCN), a novel approach for few-shot out-of-distribution detection that requires only minimal labeled samples. The method leverages CLIP's vision-language capabilities with learnable textual prompts to distinguish between in-distribution and outlier samples, advancing practical AI safety applications.
Researchers introduced PhyWorldBench, a comprehensive benchmark that evaluates text-to-video generation models on their ability to simulate real-world physics accurately. Testing 12 state-of-the-art models across 1,050 prompts, the study reveals significant gaps in how current AI video generators handle physical phenomena, from basic object motion to complex interactions, while also introducing novel evaluation methods using multimodal language models.
Researchers demonstrate that overlaying coordinate grids on chart images significantly improves multimodal LLM accuracy for data extraction tasks, reducing error rates from 25.5% to 19.5%. This spatial priming approach outperforms semantic methods like Chain-of-Thought prompting, suggesting that explicit spatial context is more effective than high-level semantic guidance for current-generation vision-language models.
SkillLens introduces a hierarchical framework for organizing and reusing skills in LLM agents at multiple granularity levels, reducing computational costs while maintaining relevance. The system retrieves and adapts skills selectively rather than injecting entire skill blocks, achieving measurable performance gains on benchmark tasks.
Researchers introduce AgentPSO, a framework that evolves multi-agent reasoning skills in large language models using particle swarm optimization principles. Rather than relying on static agents or inference-time debate, the system enables agents to iteratively improve their reasoning capabilities through self-reflection and collective learning, demonstrating improved performance and cross-benchmark transferability without modifying underlying model parameters.
A new arXiv paper argues that optimizing how language represents tasks—rather than scaling model size—is crucial for advancing LLM intelligence. The research demonstrates that deliberate language representation design can yield substantial performance improvements without modifying model parameters, supported by controlled experiments showing how different linguistic framings of identical tasks trigger different internal feature activations.
CodeClinic introduces a benchmark for evaluating whether large language model agents can autonomously generate clinical skills rather than relying on pre-built tool libraries. The research demonstrates that an offline autoformalization pipeline converting clinical guidelines into Python libraries improves consistency and reduces token usage by 40% compared to zero-shot code generation.
Researchers from UTS achieved second place in a psychological defense mechanism classification competition using a multi-agent AI system that identifies defense patterns through absence-based reasoning rather than presence detection. The system combines Gemini 2.5 agents with fine-tuned Qwen models to achieve an F1 score of 0.406, addressing critical biases in minority class prediction through structured ensemble methods.
LLARS is an open-source platform designed to streamline collaboration between domain experts and software developers in building LLM-based systems. The tool integrates prompt engineering, batch generation, and hybrid evaluation into a unified workflow, with validation from domain experts confirming significant time savings and improved interdisciplinary teamwork.
