Aging of Prompt Engineering Techniques Across LLM Versions

• URL: http://arxiv.org/abs/2608.24641v1
• Date: Tue, 25 Aug 2026 14:53:12 GMT
• Title: Aging of Prompt Engineering Techniques Across LLM Versions
• Abstract summary: We show that prompt engineering "ages" in a model-family-specific way.<n>Newer GPT models exhibit diminishing or even negative marginal gains from structured prompting.<n>Our results imply that effective prompting strategies must be adapted per model family and generation.
• Score: 1.1215180462528551
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Prompt engineering and prompt engineering techniques (PETs) have become an integral part of software engineering for AI systems. However, new LLMs are released frequently and it remains unclear how the effectiveness of prompt engineering techniques changes across successive generations of Large Language Models (LLMs). To this end, we conduct a partial replication of the study by Khojah et al. (2025). We evaluate five techniques - Zero-Shot, Few-Shot, Chain-of-Thought (CoT), Contrastive Chain-of-Thought (CCoT), and an adapted version of Program-of-Thought (PoT) - on six instruction-tuned models grouped into three version pairs: GPT-3.5-Turbo/GPT-4o, Qwen2 7B Instruct/Qwen2.5 7B Instruct, and Mistral-7B-Instruct/Mistral-Large. We use a cleaned subset of the CodePromptEval dataset with 218 context-rich Python functions and 19,620 total generations assessed via pass@k-based functional correctness to evaluate model pairs on function-level code generation tasks. We show that prompt engineering "ages" in a model-family-specific way: Newer GPT models exhibit diminishing or even negative marginal gains from structured prompting, suggesting that instruction-following and reasoning scaffolds are increasingly internalized, whereas Qwen models continue to benefit substantially from Few-Shot and CCoT. Mistral models show mixed behavior with persistent gains from CCoT but attenuated benefits from CoT and PoT. Our results imply that effective prompting strategies must be adapted per model family and generation rather than transferred unchanged. This motivates future work on adaptive, model-aware prompting and broader, multi-dimensional code quality evaluation.

Related papers

• The Missing Piece in Pre-trained Model Evaluation: Reward-Guided Decoding Unlocks Task-Oriented Behavior Without Parameter Updates [31.50160357699629]
  Energy-Based Decoding is a training-free, reward-guided framework for activating task-oriented behaviors from frozen pre-trained models.<n>We show that EBD shifts base-model outputs toward more instructionfollowing behavior, increasing behavioral similarity to post-trained counterparts.
  arXiv  Detail & Related papers  (2026-05-27T06:25:00Z)
• MERGETUNE: Continued fine-tuning of vision-language models [77.8627788911249]
  Fine-tuning vision-language models (VLMs) often leads to catastrophic forgetting of pretrained knowledge.<n>We introduce a novel paradigm, continued fine-tuning (CFT), which seeks to recover pretrained knowledge after a zero-shot model has already been adapted.
  arXiv  Detail & Related papers  (2026-01-15T15:15:53Z)
• GLUE: Generative Latent Unification of Expertise-Informed Engineering Models [3.005158583027536]
  We introduce Generative Latent Unification of Expertise-Informed Engineering Models (GLUE)<n>GLUE orchestrates pre-trained, frozen subsystem generators while enforcing system-level feasibility, optimality, and diversity.<n>On a UAV design problem with five coupling constraints, we find that data-driven approaches yield diverse, high-performing designs but require large datasets to satisfy constraints reliably.
  arXiv  Detail & Related papers  (2025-12-22T15:23:19Z)
• Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines [74.42485647685272]
  We focus on Generative Masked Language Models (GMLMs) We train a model to fit conditional probabilities of the data distribution via masking, which are subsequently used as inputs to a Markov Chain to draw samples from the model. We adapt the T5 model for iteratively-refined parallel decoding, achieving 2-3x speedup in machine translation with minimal sacrifice in quality.
  arXiv  Detail & Related papers  (2024-07-22T18:00:00Z)
• SOEN-101: Code Generation by Emulating Software Process Models Using Large Language Model Agents [50.82665351100067]
  FlowGen is a code generation framework that emulates software process models based on multiple Large Language Model (LLM) agents. We evaluate FlowGenScrum on four benchmarks: HumanEval, HumanEval-ET, MBPP, and MBPP-ET.
  arXiv  Detail & Related papers  (2024-03-23T14:04:48Z)
• Automating Human Tutor-Style Programming Feedback: Leveraging GPT-4 Tutor Model for Hint Generation and GPT-3.5 Student Model for Hint Validation [25.317788211120362]
  We investigate the role of generative AI models in providing human tutor-style programming hints. Recent works have benchmarked state-of-the-art models for various feedback generation scenarios. We develop a novel technique, GPT4Hints-GPT3.5Val, to push the limits of generative AI models.
  arXiv  Detail & Related papers  (2023-10-05T17:02:59Z)
• Stay on topic with Classifier-Free Guidance [57.28934343207042]
  We show that CFG can be used broadly as an inference-time technique in pure language modeling. We show that CFG improves the performance of Pythia, GPT-2 and LLaMA-family models across an array of tasks.
  arXiv  Detail & Related papers  (2023-06-30T17:07:02Z)
• On Robustness of Finetuned Transformer-based NLP Models [11.063628128069736]
  We characterize changes between pretrained and finetuned language model representations across layers using two metrics: CKA and STIR. GPT-2 representations are more robust than BERT and T5 across multiple types of input perturbations. This study provides valuable insights into perturbation-specific weaknesses of popular Transformer-based models.
  arXiv  Detail & Related papers  (2023-05-23T18:25:18Z)
• An automatically discovered chain-of-thought prompt generalizes to novel models and datasets [4.693905948827508]
  Chain-of-thought (CoT) reasoning capabilities promise to improve performance and explainability of large language models (LLMs) We compare different reasoning strategies induced by zero-shot prompting across six recently released LLMs. Our findings demonstrate that while some variations in effectiveness occur, gains from CoT reasoning strategies remain robust across different models and datasets.
  arXiv  Detail & Related papers  (2023-05-04T15:07:20Z)
• CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning [92.36705236706678]
  "CodeRL" is a new framework for program synthesis tasks through pretrained LMs and deep reinforcement learning. During inference, we introduce a new generation procedure with a critical sampling strategy. For the model backbones, we extended the encoder-decoder architecture of CodeT5 with enhanced learning objectives.
  arXiv  Detail & Related papers  (2022-07-05T02:42:15Z)
• Unifying Language Learning Paradigms [96.35981503087567]
  We present a unified framework for pre-training models that are universally effective across datasets and setups. We show how different pre-training objectives can be cast as one another and how interpolating between different objectives can be effective. Our model also achieve strong results at in-context learning, outperforming 175B GPT-3 on zero-shot SuperGLUE and tripling the performance of T5-XXL on one-shot summarization.
  arXiv  Detail & Related papers  (2022-05-10T19:32:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.