The Instruction Gap: LLMs get lost in Following Instruction

• URL: http://arxiv.org/abs/2601.03269v1
• Date: Fri, 19 Dec 2025 15:27:52 GMT
• Title: The Instruction Gap: LLMs get lost in Following Instruction
• Authors: Vishesh Tripathi, Uday Allu, Biddwan Ahmed,
• Abstract summary: Large Language Models (LLMs) have shown remarkable capabilities in natural language understanding and generation.<n>This study presents a comprehensive evaluation of 13 leading LLMs across instruction compliance, response accuracy, and performance metrics.
• Score: 1.4018975578160686
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Large Language Models (LLMs) have shown remarkable capabilities in natural language understanding and generation, yet their deployment in enterprise environments reveals a critical limitation: inconsistent adherence to custom instructions. This study presents a comprehensive evaluation of 13 leading LLMs across instruction compliance, response accuracy, and performance metrics in realworld RAG (Retrieval-Augmented Generation) scenarios. Through systematic testing with samples and enterprise-grade evaluation protocols, we demonstrate that instruction following varies dramatically across models, with Claude-Sonnet-4 and GPT-5 achieving the highest results. Our findings reveal the "instruction gap" - a fundamental challenge where models excel at general tasks but struggle with precise instruction adherence required for enterprise deployment. This work provides practical insights for organizations deploying LLM-powered solutions and establishes benchmarks for instruction-following capabilities across major model families.

Related papers

• Knowledge without Wisdom: Measuring Misalignment between LLMs and Intended Impact [3.437656066916039]
  LLMs increasingly excel on AI benchmarks, but doing so does not guarantee validity for downstream tasks.<n>This study evaluates the performance of leading foundation models with out-of-distribution tasks of the teaching and learning of schoolchildren.
  arXiv  Detail & Related papers  (2026-03-01T03:05:46Z)
• How well do Large Language Models Recognize Instructional Moves? Establishing Baselines for Foundation Models in Educational Discourse [0.15469452301122177]
  Large language models (LLMs) are increasingly adopted in educational technologies for a variety of tasks.<n>We compared six LLMs to estimate their baseline performance on a simple but important task: classifying instructional moves in classroom transcripts.<n>We found that while zero-shot performance was moderate, providing comprehensive examples significantly improved performance for state-of-the-art models.
  arXiv  Detail & Related papers  (2025-12-22T22:08:32Z)
• Benchmarking Large Language Models for Personalized Guidance in AI-Enhanced Learning [4.990353320509215]
  Large Language Models (LLMs) are increasingly envisioned as intelligent assistants for personalized learning.<n>This study presents an empirical comparison of three state-of-the-art LLMs on a tutoring task simulating a realistic learning setting.
  arXiv  Detail & Related papers  (2025-09-02T14:21:59Z)
• Beyond In-Context Learning: Aligning Long-form Generation of Large Language Models via Task-Inherent Attribute Guidelines [71.14354526117958]
  In-context learning (ICL) is an important yet not fully understood ability of pre-trained large language models (LLMs)<n>We present LongGuide, which efficiently generates two parallel streams of guidelines capturing task language and format properties.<n>LongGuide automatically selects the best combination of guidelines, improving both strong open- and closed-source LLMs by over 5% in both zero- and few-shot settings.
  arXiv  Detail & Related papers  (2025-06-02T02:35:24Z)
• Meeseeks: A Feedback-Driven, Iterative Self-Correction Benchmark evaluating LLMs' Instruction Following Capability [21.96694731466089]
  We introduce Meeseeks, a fully automated instruction-following benchmark equipped with an integrated feedback mechanism.<n>Meeseeks identifies erroneous components in model responses and provides corresponding feedback accurately, thereby iteratively guiding the model toward self-correction.<n>We conducted comprehensive analysis from both macro and instance levels, uncovering numerous common issues prevalent in current state-of-the-art models.
  arXiv  Detail & Related papers  (2025-04-30T13:28:19Z)
• The Inherent Limits of Pretrained LLMs: The Unexpected Convergence of Instruction Tuning and In-Context Learning Capabilities [51.594836904623534]
  We investigate whether instruction-tuned models possess fundamentally different capabilities from base models that are prompted using in-context examples.<n>We show that the performance of instruction-tuned models is significantly correlated with the in-context performance of their base counterparts.<n>Specifically, we extend this understanding to instruction-tuned models, suggesting that their pretraining data similarly sets a limiting boundary on the tasks they can solve.
  arXiv  Detail & Related papers  (2025-01-15T10:57:55Z)
• Enhancing and Assessing Instruction-Following with Fine-Grained Instruction Variants [28.691691883519542]
  We introduce a technique that decomposes complex instructions into simpler sub-components, modifies these, and reconstructs them into new variants. Based on DeMoRecon, we developed the FGIV dataset which contains fine-grained instruction variants of 1,773 seed instructions. Our findings show that LLMs fine-tuned with FGIV will gain significant performance boost on both ours and commonly used instructions-following benchmarks.
  arXiv  Detail & Related papers  (2024-06-17T08:08:11Z)
• PPTC-R benchmark: Towards Evaluating the Robustness of Large Language Models for PowerPoint Task Completion [96.47420221442397]
  We construct adversarial user instructions by attacking user instructions at sentence, semantic, and multi-language levels. We test 3 closed-source and 4 open-source LLMs using a benchmark that incorporates robustness settings. We find that GPT-4 exhibits the highest performance and strong robustness in our benchmark.
  arXiv  Detail & Related papers  (2024-03-06T15:33:32Z)
• InFoBench: Evaluating Instruction Following Ability in Large Language Models [57.27152890085759]
  Decomposed Requirements Following Ratio (DRFR) is a new metric for evaluating Large Language Models' (LLMs) ability to follow instructions. We present InFoBench, a benchmark comprising 500 diverse instructions and 2,250 decomposed questions across multiple constraint categories.
  arXiv  Detail & Related papers  (2024-01-07T23:01:56Z)
• Benchmarking Large Language Models on Controllable Generation under Diversified Instructions [34.89012022437519]
  Large language models (LLMs) have exhibited impressive instruction-following capabilities. It is still unclear whether and to what extent they can respond to explicit constraints that might be entailed in various instructions. We propose a new benchmark CoDI-Eval to evaluate LLMs' responses to instructions with various constraints.
  arXiv  Detail & Related papers  (2024-01-01T07:35:31Z)
• Benchmarking Generation and Evaluation Capabilities of Large Language Models for Instruction Controllable Summarization [132.25202059478065]
  We benchmark large language models (LLMs) on instruction controllable text summarization. Our study reveals that instruction controllable text summarization remains a challenging task for LLMs.
  arXiv  Detail & Related papers  (2023-11-15T18:25:26Z)
• Can Large Language Models Understand Real-World Complex Instructions? [54.86632921036983]
  Large language models (LLMs) can understand human instructions, but struggle with complex instructions. Existing benchmarks are insufficient to assess LLMs' ability to understand complex instructions. We propose CELLO, a benchmark for evaluating LLMs' ability to follow complex instructions systematically.
  arXiv  Detail & Related papers  (2023-09-17T04:18:39Z)
• Metacognitive Prompting Improves Understanding in Large Language Models [12.112914393948415]
  We introduce Metacognitive Prompting (MP), a strategy inspired by human introspective reasoning processes. We conduct experiments on four prevalent Large Language Models (LLMs) across ten natural language understanding (NLU) datasets. MP consistently outperforms existing prompting methods in both general and domain-specific NLU tasks.
  arXiv  Detail & Related papers  (2023-08-10T05:10:17Z)

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.