Order Matters: Investigate the Position Bias in Multi-constraint Instruction Following

• URL: http://arxiv.org/abs/2502.17204v2
• Date: Mon, 03 Mar 2025 06:29:31 GMT
• Title: Order Matters: Investigate the Position Bias in Multi-constraint Instruction Following
• Authors: Jie Zeng, Qianyu He, Qingyu Ren, Jiaqing Liang, Yanghua Xiao, Weikang Zhou, Zeye Sun, Fei Yu,
• Abstract summary: Real-world instructions with multiple constraints pose a significant challenge to existing large language models (LLMs)<n>We quantitatively measure the difficulty distribution of the constraints by a novel Difficulty Distribution Index (CDDI)<n>We find that LLMs are more performant when presented with the constraints in a hard-to-easy'' order.
• Score: 39.114513139453756
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Real-world instructions with multiple constraints pose a significant challenge to existing large language models (LLMs). An observation is that the LLMs exhibit dramatic performance fluctuation when disturbing the order of the incorporated constraints. Yet, none of the existing works has systematically investigated this position bias problem in the field of multi-constraint instruction following. To bridge this gap, we design a probing task where we quantitatively measure the difficulty distribution of the constraints by a novel Difficulty Distribution Index (CDDI). Through the experimental results, we find that LLMs are more performant when presented with the constraints in a ``hard-to-easy'' order. This preference can be generalized to LLMs with different architecture or different sizes of parameters. Additionally, we conduct an explanation study, providing an intuitive insight into the correlation between the LLM's attention and constraint orders. Our code and dataset are publicly available at https://github.com/meowpass/PBIF.

Related papers

• Time's Up! An Empirical Study of LLM Reasoning Ability Under Output Length Constraint [20.685932824324446]
  We investigate whether and how the reasoning abilities of Large Language Models (LLMs) remain effective under real-world latency constraints. Specifically, we test more than 25 LLMs on common reasoning datasets under a wide range of output length budgets. The results have demonstrated several interesting findings regarding the budget-aware LLM reasoning.
  arXiv  Detail & Related papers  (2025-04-19T16:32:28Z)
• WildIFEval: Instruction Following in the Wild [4.5214954812238295]
  We introduce WildIFEval - a large-scale dataset of 12K real user instructions with diverse, multi-constraint conditions. Unlike prior datasets, our collection spans a broad lexical and topical spectrum of constraints, in natural user prompts. We categorize these constraints into eight high-level classes to capture their distribution and dynamics in real-world scenarios.
  arXiv  Detail & Related papers  (2025-03-09T12:06:29Z)
• Preference Leakage: A Contamination Problem in LLM-as-a-judge [69.96778498636071]
  Large Language Models (LLMs) as judges and LLM-based data synthesis have emerged as two fundamental LLM-driven data annotation methods.<n>In this work, we expose preference leakage, a contamination problem in LLM-as-a-judge caused by the relatedness between the synthetic data generators and LLM-based evaluators.
  arXiv  Detail & Related papers  (2025-02-03T17:13:03Z)
• Divide-Verify-Refine: Aligning LLM Responses with Complex Instructions [33.18076221854853]
  LLMs struggle to follow complex instructions with multiple constraints. Recent studies show that LLMs, particularly open-source models, struggle to follow complex instructions with multiple constraints. We propose the Divide-Verify-Refine (DVR) framework with three steps. We show that the framework significantly improves performance, doubling LLama3.1-8B's constraint adherence on instructions with 6 constraints.
  arXiv  Detail & Related papers  (2024-10-16T04:01:55Z)
• Prompt Recursive Search: A Living Framework with Adaptive Growth in LLM Auto-Prompting [22.025533583703126]
  We propose a novel Prompt Recursive Search (PRS) framework for large language models (LLMs) PRS framework incorporates an assessment of problem complexity and an adjustable structure, ensuring a reduction in the likelihood of errors. Compared to the Chain of Thought (CoT) method, the PRS method has increased the accuracy on the BBH dataset by 8% using Llama3-7B model, achieving a 22% improvement.
  arXiv  Detail & Related papers  (2024-08-02T17:59:42Z)
• From Distributional to Overton Pluralism: Investigating Large Language Model Alignment [82.99849359892112]
  We re-examine previously reported reductions in response diversity post-alignment. Our analysis suggests that an apparent drop in the diversity of responses is largely explained by quality control and information aggregation. Findings indicate that current alignment techniques capture but do not extend the useful subset of assistant-like base LLM behavior.
  arXiv  Detail & Related papers  (2024-06-25T16:32:33Z)
• Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction [75.25114727856861]
  Large language models (LLMs) tend to suffer from deterioration at the latter stage ofSupervised fine-tuning process. We introduce a simple disperse-then-merge framework to address the issue. Our framework outperforms various sophisticated methods such as data curation and training regularization on a series of standard knowledge and reasoning benchmarks.
  arXiv  Detail & Related papers  (2024-05-22T08:18:19Z)
• 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)
• The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning [61.68787689234622]
  A recent study, LIMA, shows that using merely 1K examples for alignment tuning can achieve significant alignment performance as well. This raises questions about how exactly the alignment tuning transforms a base LLM. We show that the gap between tuning-free and tuning-based alignment methods can be significantly reduced through strategic prompting.
  arXiv  Detail & Related papers  (2023-12-04T00:46:11Z)
• FollowBench: A Multi-level Fine-grained Constraints Following Benchmark for Large Language Models [79.62191017182518]
  FollowBench is a benchmark for Fine-grained Constraints Following Benchmark for Large Language Models. We introduce a Multi-level mechanism that incrementally adds a single constraint to the initial instruction at each increased level. By evaluating 13 popular LLMs on FollowBench, we highlight the weaknesses of LLMs in instruction following and point towards potential avenues for future work.
  arXiv  Detail & Related papers  (2023-10-31T12:32:38Z)

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.